The real problem is that cyclists typically ride near the edge of the lane or road rather than in the center of the lane. This puts the cyclist outside of the foveal field of view for the motorist, which means that they're less likely to see them in time to avoid a collision.
Unfortunately, where I live bicycles keeping to the side of the lane is required by law. Even in situations where this isn't required, drivers can easily become aggressive towards slower vehicles such as bicycles.
Bicycles often fall through the cracks: not quite first class vehicles but not afforded the same protections as pedestrians.
Where I live there's also a law like that, but if you look a little closer you realize that it actually requires cyclists to maintain a sizable safety margin to the actual edge while keeping right.
Of course many drivers only know the general headline which leads to a lot of, uhm, "ad-hoc driver education"...
> Unfortunately, where I live bicycles keeping to the side of the lane is required by law.
In the US, most vehicle law is based on the UVC (uniform vehicle code) template. While that does contain a clause that cyclists must keep as far right as practicable, it lists a number of exceptions. One of those exceptions is a "substandard width lane"[1]. The law defines it as a lane that's too narrow for a bicycle and another vehicle to safely pass side by side within the lane.
Most general purpose lanes are 10 to 12 feet wide. A cyclist can range from 2 to 3 feet wide (depending on the type of bicycle) and require a minimum of 4 feet of operating space[2] with 5 feet preferred.
Even with the cyclist operating at far right as practicable and taking the minimum operating width into account, they take a little over 3 feet in roadway width. Given that most states have a 3 foot minimum distance passing law and the fact that most motor vehicles range from 7 to 8 feet in width (including their side mirrors) and commercial vehicles are around 10 feet wide, it's not possible to safely or legally pass a cyclist while remaining within a typical 10 to 12 foot wide lane.
You need to be quite confident to do this as it can often result in car drivers honking, shouting or overtaking recklessly and nearly clipping you. Less experienced cyclists often feel much safer hugging the right side of the lane, even though as you mention this is much more dangerous, especially the risk of being doored (I learned that lesson the hard way). Depends on the city and cycling culture a bit but thats my experience.
> You need to be quite confident to do this as it can often result in car drivers honking, shouting or overtaking recklessly and nearly clipping you.
Actually, the opposite is true. When you ride in the center of the lane, motorists will typically make a full lane change to pass you. If you ride near the edge of the lane, motorists will try to squeeze between you and the lane line.
A full lane change definitely allows for more distance between the cyclist and motorist.
I was a confident cyclist, I commuted daily and a could sprint to 30mph, could hold 20mph, and was happy holding my road position. In London, realistically traffic never really got much about 20, so I should have been golden. But the amount of cars that either tried to squeeze by me, honk or shout was huge - a daily occurrence. Even just to gain a car length or two, and always only until the traffic light,but drivers nearly always want to go as fast as their car (and the limit) allows and anything in their way is an inconvenience.
If this happened to me on a fast road bike, I think someone on a sit up and beg bike doing 7 to 12mph is going to find holding a road position even harder. And so people ride in the gutter, with no where to turn left or right when an obstacle is ahead or someone comes too close.
When I refer to riding in the center of the lane, I mean riding between the center and left tire track (or between the center and right tire track when traffic keeps to the left). Almost invariably over the last 15 years or so, I have found that overtaking drivers will change lanes[1].
When I used to ride in the right tire track position (left tire track position where traffic drives on the left), I would frequently have overtaking vehicles squeeze between me and traffic in the adjacent lane.
I ride a hybrid bike and can maintain about 15 mph on flat ground and maybe 5 to 8 mph uphill. I still take the lane regardless and will release when safe to do so.
Are these multi lane roads? In London the roads I was on are usually a single lane in either direction. So to overtake me properly they need to move into the oncoming traffic lane.
I hold a position in the centre of my lane, other than when I pull in to allow a vehicle to pass - I'll never ride in the gutter for any length of time. But if I feel I'm holding up traffic and it is safe to do so, I'll move in - but as I say in London I was rarely holding up traffic - and if I was it was only until the next traffic light which can be every 500m or even less!
If you're referring to multiple lanes for same direction traffic, then some of them are. Others only had a single lane for same direction traffic divided by a center line.
In both cases, traffic would move into the next lane (whether for same or opposite direction traffic) to overtake me or they would slow to my speed and wait for an opportunity to pass. In the latter case, if there wasn't a safe opportunity for overtaking coming up for a bit, I would release the lane by either pulling off the road and letting same direction traffic by, or if there was more space, move a little further towards the edge and let the motorist past me (re-taking the lane afterward). But on roads where there are multiple lanes for same direction traffic, I would hold my position and they would eventually change lanes to pass.
I'm pleased you live in a country that is safe for cyclists with respectful drivers. I don't feel it is that way in the UK ;) - most people aren't terrible, but most don't consider cyclists to have the same rights to the road
I'm not saying that I haven't been harrassed, but something like that might happen once every couple of years. In contrast, when I used to ride closer to the edge of the road/lane, I frequently experienced close overtakes.
Yes, ultimately infrastructure is the real solution here. Take the Netherlands for example. We interviewed folks from nearly every country on this, and the response in the Netherlands was heartwarming:
Every person we talked to had never even heard of someone being hit by a car while riding a bike (or walking). And after the interview, they interviewed a bunch of their friends/family/etc. and came back to us with those results: Still the same, not a single story of an accident.
And this all goes back to the way the Dutch design infrastructure. A ton has been written/filmed/etc about this, but here's an example: https://www.youtube.com/watch?v=XuBdf9jYj7o
Yeah... I live near a well-marked shared lane and frequently ride down it. I stay to the right-ish except when passing parked cars (which I give berth to avoid getting doored). Not just cars, but even municipal buses, get aggro despite the frequent signage.
This seems like a good example of the problems with Silicon Valley brain, where we try to solve problems by hacking around them, rather than solving the actual problem, which is poor cycling facilities, easily solved by installing separated bike lanes.
In Copenhagen or Amsterdam that's just the case. (Not literally every road, they don't have bike lanes in residential areas, but almost every major road with significant traffic has a bike lane in these cities.)
Not just Amsterdam, the whole of the Netherlands is like that. But more than that, there is the cultural aspect that cyclists and pedestrians are the number 1 priority on the road, then cars. None of these fancy hi-tech solutions work if the drivers don't care, and the laws don't protect cyclists.
Pretty much, except for small residential (or city centre) roads.
A dashed blue edge to the road means a lane separated by a kerb from cars and pedestrians. A solid blue edge means a line of paint on the road. There are also off-road paths with dashed, blue lines.
The thick, translucent lines are just signed, long-distance routes; they are likely to be mostly good, but not necessarily.
Of course you can. Unless it's low-traffic enough that it's not needed, every road here has a dedicated bike lane, and usually it's a separate path. I think it's a legal requirement to do so.
This product feels like attempting to put a technological band-aid over a poor infrastructure planning problem.
>> But you can't expect every road to have a dedicated bike lane, right?
> Of course you can.
The road has to be wide enough to accommodate a bike lane. The absolute minimum width for a bike lane is 4 feet (not including the gutter adjoining the curb).
In my town, they recently restriped a 4 lane road with a concrete raised median and sidewalks on both sides to include bike lanes. They basically narrowed the 2 general purpose traffic lanes from 12 feet to 10 feet in width and added a 4 foot wide bike lane.
With the bike lane, if my bike with a 3 foot wide trailer is centered in the bike lane, I have about 0.5 feet to each side of me. A car with a mirror span of 7.5 feet centered in the 10 foot wide general purpose lane has about 1.25 feet on each side. 1.25 + 0.5 = 1.75 feet of clearance between a car in the right lane and a cyclist in the bike lane.
With the old configuration, a cyclist centered in the 12 foot wide general purpose traffic lane would have 4.5 feet of space on each side. A car centered in the lane would have 2.25 feet of space on each side. 4.5 + 2.25 = 6.75 feet of space between the car in the left lane and a cyclist in the right lane.
In that case, having a bike lane substantially decreased the possible passing distance between cars and bicycles. If they wanted to add the bike lane, then the road should have been at least 2 to 3 feet wider than it is now.
> every road here has a dedicated bike lane, and usually it's a separate path.
Those lead to issues at intersections where turning vehicles or vehicles entering the road may not see you in time when you're crossing.
I mean, you can say it doesn't work all you like, but the reality is that it does, and there is concrete (well, asphalt) evidence right outside my window and all over the rest of the city and country that it works and works well. It just requires the will to design things properly.
Separated paths improve safety, they don't cause issues at intersections that I've noticed (and I've been biking through them every days for years), probably because things were designed well to account for this. It's one of those situations where good design becomes mostly invisible so it's hard to spot.
> I mean, you can say it doesn't work all you like, but the reality is that it does
Not really. This document describes the crashes involving cyclists in a bidirectional separated bike path[1] which for a couple of years was having close to one crash per month.
That looks to my mostly uneducated eye like a badly designed intersection. You can't just slap a cycle lane (especially two-way) in and expect it to work, that's a poor solution.
Compare to here, https://goo.gl/maps/F1mufzKACT3XpDZa6, just as a fairly random example. You see that the car traffic is forced, via traffic calming (i.e. the raised footpath) to slow, has obvious lower priority than bikes and pedestrians, and can't really proceed without being strongly encouraged to look first. A couple more interesting samples at the bottom that are designed with safety in mind, and in my experience work well.
Taking an existing road, slapping a bike line on one side, and calling it a useful test is not really effective. This is why I mentioned design is important.
What is interesting from your PDF (which I just skimmed, so don't know if it's representative) is that the number of people cycling increased, so it looks like people felt it was safer even if it really wasn't. I suspect that if you managed to get that up to real numbers of people cycling, then drivers would naturally become a lot more aware also.
Because at least we know it'll have a lifetime longer than a few years when said company feels like bringing out a new, incompatible version, or gets bought out by someone who then discontinues the product.
Improved infrastructure helps everyone in a much more effective manner than a technological solution needing to be reliably used by hundreds of thousands of individuals.
It also is well-intended, but also an example of blaming the cyclist for being there (a lot less blaming the cyclist than saying “people who ride bikes don’t get hit by cars – that they hit the cars.”, but still blaming). Why would cyclists have to buy this kind of hardware to stay alive on the road?
I also don’t see how this would do the “prevent accidents” part of “could detect when such collisions were going to occur to take action and prevent accidents”.
Who’s going to take action here? If it’s the cyclist, how are they going to tell them what to do, and how are the cyclists going to do the right thing, given, likely, less than a second to act? (If you’re going to warn them earlier, I would hope there will be lots of false alarms for cars that see you, but don’t take action yet. Or am I mistaken in understanding what kind of war zone cyclists live in in the USA? We aren’t talking about cyclists being pumped up with adrenaline all the time, are we?)
I would guess a rear-view camera or, way simpler, rear-view mirror would be the better and easier way to increase situational awareness of cyclists?
> easily solved by installing separated bike lanes.
In environment where there frequent intersections, these actually lead to more problems for cyclists at those intersections due to risks of collisions from:
1. right-hooks where traffic makes a right turn and crosses the path of the cyclist
2. left-crosses where left turning traffic crosses the path of the cyclist
3. drive-outs where traffic pulling from a side-street or driveway crosses the path of a cyclist.
These are much less of an issue if the cyclist rides in the center of the right-most general purpose lane since motorists will be looking for traffic in those locations.
The hardware/firmware/software/AI necessary to make such a device wasn't available when we started this though, so we had to make it first. And we're releasing it with OpenCV. See here for details:
https://opencv.org/opencv-spatial-ai-competition/
So the bike-safety idea is not a product yet, but the underlying platform is now (and is going live on KickStarter soon).
I'm not sure I understand what the product is and what it does - it seems more like a technology looking for a market.
Is this a product that end users (car owners) could buy? How would it be integrated with the car? As a driver who doesn't care about cyclists (as those are the ones causing most accidents), would you buy a specific device just for avoiding cyclists?
Is it a product car manufacturers have to integrate in the car? Shouldn't this have a wider target than cyclists? How does it compare to the existing collision avoidance systems? If it's an optional safety system just for detecting cyclists, how do you sell it to the car buyer?
> So the final product is a multi-camera, rear-facing bike-light, which uses object detection and depth mapping to track the cars behind you
It's for the cyclist.
I'm both a car driver and a cyclist. I wonder if I want something like this on my bike and be unnecessarily scared by false positives (cars do get close to bicycles all the times and what can I do anyway?) or on my car just in case I don't notice a bicycle with no lights at night (plenty of them here.)
Ok, so it's a camera implementation of the Garmin Varia bike radar, with special functions to scare away drivers. Yes, as you say, a false positive would be really annoying for both cyclists and the drivers, as they can blind them and actually create an accident (what happens when you suddenly see a powerful flashing light during the night? flashing lights on bikes are annoying enough as they are, without an extra-powerful setting).
And also if the horn goes off, would that make agri drivers more so? I used to have an air horn on my bike - it sounded like a lorry horn, and the issue was that drivers heard it and would stop but would normally assume it wasn't from me, and would carry on.
That's a really good point, it would definitely escalate aggressive drivers.
I've had situations where I rode past someone (they were stopped in traffic), then they passed me again down the road (this repeated a couple of times) and every time I rolled past them, they'd descend further into apoplectic rage, I can't imagine flashing lights is going to help.
I can imagine: they think they paid their car 100 times more than our bicycles (actually maybe only 10), they know they could go 10 times as fast as a bicycle (maybe only 5) and yet they are stuck in the traffic, go as slow as us and won't be at home much earlier (or much later.)
When it happens to me to be a car driver I think that at least I'm enjoying heating or air conditioning but I'd like to be on my bicycle at least until I get out of the city.
Seems like an overall well-intentioned founder story. Wondering if there are Luxonis people out there who can address the below?
Would like to give the benefit of the doubt here to the CEO and Intel, but being weary of Intel’s past “hacker friendly” blowharding and subsequent total abandonment of the hacker community with the Edison, I do wonder whether Intel is again playing the mafioso; the wording of this quote from the Luxonis CEO comes off, to me at least, as kissing the ring,
> “DepthAI, powered by Intel Movidius Myriad technology, is making it tractable.”
Also wonder about the nature of the open source software the article refers to,
> The Luxonis CEO said that the SoM works with a wealth of code that the company has open-sourced through an MIT license.
Is the software running on the Myriad X itself open source, or just the SDK interfacing with it from the host machine? If the former, pretty amazing they were able to get Intel to agree to this, as I imagine the chip details are under strict NDA. If the latter, it seems pretty disingenuous to refer to the SOM software as “open source,” at least not with a footnote that the most interesting part (the spatial depth calculation) is not, presuming that’s done on the Myriad X.
Thanks for the kind words. Brandon the founder here. :-)
So although I could see how that quote could appear like that, the quote is not kissing the ring. We scoured the whole semiconductor start up scene to find a chip that could be used for this.
There are only 2 chips (as I'm aware, as of this writing) that can be used in such a way:
1. Intel Movidius Myriad X
2. Inuitive NU4000AI.
And the Inuitive, until super recently (i.e. 1.5 years after we have already built hardware off the Myriad X) was not ready to be used (had tape-out issues at the fab).
So this made the Myriad X the only chip that (1) had the capabilities needed and (2) was actually available and in production.
In terms of our open-source nature, the latter is what we've implemented - where we have closed source binaries running on the Myriad X - which then have a slew of open-source counterparts on the host. Sorry if this came off disingenuous. Do you have advice on how to phrase it in a better way?
The above includes open source hardware, core capabilities in C++ which can be cross-compiled for various hosts, and open-source training notebooks and use-case examples.
So the goal of open-sourcing all that we can (including hardware) is to enable folks who have their own applications to leverage this, modify it, use it, etc. w/out having to even talk to us. We cannot open-source the code that runs on the Myriad X however, as then we wouldn't have a way to monetize, and would have to give up on the mission. (That, and we're also not allowed to.)
"So the final product is a multi-camera, rear-facing bike-light, which uses object detection and depth mapping to track the cars behind you - and know their x, y, and z position relative to you (and the x, y, and z of their edges, which is important).
What does it do with this information?
It's normally blinking/patterning like a normal bike light, and if it detects danger, it does one of two actions, and sometimes both:
1. Starts flashing an ultra-bright strobe - to try to get the driver's attention
2. Initiates haptic feedback - to let the person riding the bike that they're in danger
3. Honks a car horn. This is if all else fails, and only happens if you are in certain danger"
This sounds like a great idea in theory but in practice my guess is that its too little (this won't help cars pulling out of junctions into bikes and many other dangerous scenarios) and too late (how much time will the driver have to see the strobe/hear the horn and take evasive action).
The Radar powered Garmin Varia (https://buy.garmin.com/en-GB/GB/p/601468) is probably simpler and cheaper and I'd guess more likely to work reliably in bad lighting conditions.
But a good set of continuously flashing ultra bright strobe lights, high visibility/reflective clothing, and bike lanes or driver education is by far superior.
Thanks for the comments! So the radar solution is great for touring riders, where one doesn't see a car for half an hour or a couple hours. It mentally wakes them up to the presence of a car.
It's not intended for use in commuting though... as in commuting you don't care about the 500 cars that will pass you safely on your ride (and you don't want 500 alerts), you care about the 1 that is on a trajectory to kill you.
So with computer vision you can tell that trajectory. You can also with radar, but only with super-expensive radars ($100k+), wheras you can do this w/ computer vision for $50 or so.
So it's like the radar-based Garmin RTL5xx (née Backtracker) but with the added benefit of "hit or pass" lane information and all the warning escalation you can do with that. I'm generally extremely sceptical of technical solutions to driver inventiveness (they will eventually out-risk-compensate everything you throw at them, it's a one-sided arms race) but they might be on to something here. The Garmin is already surprisingly popular and proper use of lateral information would greatly add to the appeal. A contender for the annual Garmin Fitness acquisition 2021 perhaps?
edit: Still, I'd be afraid that falls from startled reactions by the cyclists themselves might in the end cause more bloodshed than the extremely rare rear-ender it's supposed to prevent. A car preparing a well timed overtake (close in fast inside the lane to sweep left just after oncoming traffic passes) will be a false positive you can't avoid without directly guessing intention and skill of the driver. It will surely be a net-positive in some communities with particularly low cyclist awareness amongst drivers but not necessarily in others.
Thanks for the comments usrusr! And great points. So one of the things we did early-on is to prototype the whole flow to see if you could indeed get incredibly low false-positives. The key ended up being the combination of disparity depth (which gives over 1 million depth points) and neural inference so that the system can know incredibly granularly (i.e. within inches) the edges of things and their predictive trajectories.
So at the outset of this effort (back in 2017) such a computer-vision based device that was inexpensive and embeddable was impossible. The Movidius Myriad X came out, which has all the requisite CV/AI processing to make such a thing solvable... but there was no hardware/firmware/software/AI ecosystem which allowed producing such embedded spatial AI problems.
So the `why` of this is it allows others to solve problems on embedded systems which were previously intractable. And there are a ton of them... the number of applications we've seen has blown our mind.
I see in Germany older folks often using a mirror on the left handlebar. Once I borrowed such bicycle, it was somehow weird, but I got used to that. Such device identifies cars and dangerous situation very well. Zero false positives, works pretty well in the night too. No batteries needed and costs between 5 and 15€.
Yes, mirrors are a fantastic solution. And for getting the attention of the distracted driver (as sometimes the dynamics of the bike preclude maneuvering out of the way in time), LOUD bicycle is a great solution: https://www.kickstarter.com/projects/lansey/loud-bicycle-car...
It works really well, but requires the cyclist to assess the risk and react. Something like this, that could inform both myself and the driver of an impending collision, would be (literally) a life saver.
Yes, totally agree on V2X. That’s another hard infrastructure problem though… as it requires so many parties to adopt it (and implement it properly) for it to really help. We did talk to a bunch of V2X folks on this though to see how we could make an integration happen… but no traction so far.
Thanks for the reassuring words sunpazed! Yes, that’s exactly right... where the Garmin product is great for alerting the rider there exists a vehicle behind them - which is SUPER useful for touring riders - this alerts the rider and the driver of an impending collision. Which is what makes it applicable to bicycle commuting, where say 500 safe passes occur per commute, and you only care about the one that’s going to hit you.
Radar works great for presence, but is extremely expensive compared to computer vision for determining exact positioning and 3D trajectory. And computer vision (specifically the combination of stereo CV/AI) is actually super-good at this, and can see quite granularly (i.e. 1 million depth points every 60 times a second out to ~20 meters).
So you're telling potentially distracted drivers that hey, it's OK to be distracted, the AI will take care of that for you? Like with Tesla's auto-pilot, it doesn't matter what the product manual says - too many drivers will interpret it to mean you can pay less attention and the computer will take up the slack for you.
Luxonis-Brandon here.
So the distracted driver actually wouldn't be aware that the person riding the bike has this installed... unless the driver is paying super-close attention, in which case the device is not needed.
So unlink the Tesla example, where the driver knows his vehicle has this safety feature... the driver does not know the biker has this feature.
My experience of cycling in the UK is many cyclists already go to great lengths to make themselves visible, and it makes no difference at all to close passing, etc. Hi-viz clothing, bright strobing lights - drivers act like you're not there.
It also makes drivers lazy - having nearly been knocked off my bike, the driver stopped and berated me for not wearing hi-viz, despite the bright lights I was using. Not impossible, therefore, that I also get berated for not using your device to warn me about drivers. There have been attempts to do this in the UK already with apps which notify drivers of cycists, with the app maker saying it would be the cyclists' fault if they were in an accident and weren't themselves using the app. (This was repudiated by the Advertising Standards Authority, but still.) This is similar to the all-too-frequent reporting about "A cyclist was crushed to death by a truck. Police said the victim was not wearing a helmet." Even the helmet manufacturers don't say it will stop a 20-ton vehicle from killing you :(
As Chris Boardman says about the helmet debate, it's the wrong emphasis. The issue is safety on our roads and drivers being careful around cyclists. Also, cyclists being predictable and riding safely. An electronic device will make little difference without that, and I feel it could make things worse.
Also, I find a mirror on the handlebars, a life-saver check when turning, and listening for engine noise to be pretty good at alerting me to drivers. Not sure electronics would improve this.
Agreed that the ultimate solution is infrastructure.
That's what we harp as well (Luxonis-Brandon here). And we actually see our goal as means to get there... to do what we can (as technologists) to try to keep people who ride bikes safe, keep people riding bikes, and keep there being demand for the infrastructure going.
If everyone is too afraid to ride bikes like my circle is (reminder: 1 of my friend was killed, and 3x got broken femurs, broken hips, and 1 of those got a traumatic brain injury), then there won't be demand for such infrastructure.
And with lower demand, it's harder for there to be incentive to make the infrastructure, so the infrastructure may actually get worse - making it even less safe - and then resulting in even fewer people riding bikes.
And there's another constraint: I'm an Electrical Engineer and such tech solution are all I’m good at. I don't know how to make roads with bike lanes. And more importantly, I don't know how to influence politicians to make these roads. (And my wiring is such that I'd never be good at it, no matter how hard I try.)
That, and there are ~4 million miles of road in the US. So if we put the whole economy onto making new roads as fast as we could, we'd be a decade or more realistically probably two or more decades out until we had the new roads with dedicated bike lanes.
This is what The Netherlands did, by the way, and they accomplished it (in a comparatively small country) in about 2 decades. And this accomplishment is why there exists absolutely zero car-bike accidents in the Netherlands (and much lower car-car accidents too).
So our goal is to try to help on the demand side. If there isn't demand for safe infrastructure, that infrastructure won't come. And there won't be demand if people are too afraid to ride bikes because of deaths and injuries of their friends/family.
So we want to try to use technology as a bandaid until the infrastructure is there, to try to keep people safe, and to also then help drive making infrastructure (by making demand for the infrastructure).
Speaking of which, this device quantizes risk. So when used on a bike, it would allow mapping quantized risk to bikers, which then helps to advise which are the most important routes/streets/etc. to be improved for safety. And also advises people who want to ride bikes as to which routes they should avoid.
So then when infrastructure is installed, this system can make it so the most-fatal locations get infrastructure first. So it's more prioritized and effective.
We agree that segregated infrastructure is the ultimate solution. But we're technologists and are 0% effective at lobbying for infrastructure. Having had a bunch of friends hurt and killed by distracted drivers, it's us using the skills and capabilities we have to try to make a difference.
And the core of the development is the computer vision technology which allows a whole slew of other applications, including other safety solutions. The sheet numbers of problems solvable with this sort of application-specific human level perception is crazy.
We'll be releasing the winners of stage 1 soon (please feel free to apply!), which will quantify how crazy it is to be able to approach human-level perception (what an object is, and where) for specific myopic tasks (like picking strawberries, or detecting when a car is going to run you over).
Anyway, back to it, I agree personally the solution is infrastructure (which I've tried to write in every location that covers this... but maybe failed to do in that interview). But I view this as a stepping stone to infrastructure. If people are too scared to ride bikes because of the injury and death at the hands of distracted people driving cars, then there will be no push for infrastructure (as why build infrastructure for a thing no one does?).
So the goal is to allow people to ride bikes -safer- before the infrastructure is there... and thereby help drive the infrastructure change (through supply and demand relationship).
67 comments
[ 3.8 ms ] story [ 140 ms ] threadBicycles often fall through the cracks: not quite first class vehicles but not afforded the same protections as pedestrians.
Of course many drivers only know the general headline which leads to a lot of, uhm, "ad-hoc driver education"...
In the US, most vehicle law is based on the UVC (uniform vehicle code) template. While that does contain a clause that cyclists must keep as far right as practicable, it lists a number of exceptions. One of those exceptions is a "substandard width lane"[1]. The law defines it as a lane that's too narrow for a bicycle and another vehicle to safely pass side by side within the lane.
Most general purpose lanes are 10 to 12 feet wide. A cyclist can range from 2 to 3 feet wide (depending on the type of bicycle) and require a minimum of 4 feet of operating space[2] with 5 feet preferred.
Even with the cyclist operating at far right as practicable and taking the minimum operating width into account, they take a little over 3 feet in roadway width. Given that most states have a 3 foot minimum distance passing law and the fact that most motor vehicles range from 7 to 8 feet in width (including their side mirrors) and commercial vehicles are around 10 feet wide, it's not possible to safely or legally pass a cyclist while remaining within a typical 10 to 12 foot wide lane.
[1] https://law.lis.virginia.gov/vacode/title46.2/chapter8/secti...
[2] https://njdotlocalaidrc.com/perch/resources/aashto-gbf-4-201... (page 52)
Actually, the opposite is true. When you ride in the center of the lane, motorists will typically make a full lane change to pass you. If you ride near the edge of the lane, motorists will try to squeeze between you and the lane line.
A full lane change definitely allows for more distance between the cyclist and motorist.
I was a confident cyclist, I commuted daily and a could sprint to 30mph, could hold 20mph, and was happy holding my road position. In London, realistically traffic never really got much about 20, so I should have been golden. But the amount of cars that either tried to squeeze by me, honk or shout was huge - a daily occurrence. Even just to gain a car length or two, and always only until the traffic light,but drivers nearly always want to go as fast as their car (and the limit) allows and anything in their way is an inconvenience.
If this happened to me on a fast road bike, I think someone on a sit up and beg bike doing 7 to 12mph is going to find holding a road position even harder. And so people ride in the gutter, with no where to turn left or right when an obstacle is ahead or someone comes too close.
When I refer to riding in the center of the lane, I mean riding between the center and left tire track (or between the center and right tire track when traffic keeps to the left). Almost invariably over the last 15 years or so, I have found that overtaking drivers will change lanes[1].
When I used to ride in the right tire track position (left tire track position where traffic drives on the left), I would frequently have overtaking vehicles squeeze between me and traffic in the adjacent lane.
I ride a hybrid bike and can maintain about 15 mph on flat ground and maybe 5 to 8 mph uphill. I still take the lane regardless and will release when safe to do so.
[1] https://iamtraffic.org/wp-content/uploads/2015/05/BicyclistP... (figure on page 8)
I hold a position in the centre of my lane, other than when I pull in to allow a vehicle to pass - I'll never ride in the gutter for any length of time. But if I feel I'm holding up traffic and it is safe to do so, I'll move in - but as I say in London I was rarely holding up traffic - and if I was it was only until the next traffic light which can be every 500m or even less!
If you're referring to multiple lanes for same direction traffic, then some of them are. Others only had a single lane for same direction traffic divided by a center line.
In both cases, traffic would move into the next lane (whether for same or opposite direction traffic) to overtake me or they would slow to my speed and wait for an opportunity to pass. In the latter case, if there wasn't a safe opportunity for overtaking coming up for a bit, I would release the lane by either pulling off the road and letting same direction traffic by, or if there was more space, move a little further towards the edge and let the motorist past me (re-taking the lane afterward). But on roads where there are multiple lanes for same direction traffic, I would hold my position and they would eventually change lanes to pass.
Every person we talked to had never even heard of someone being hit by a car while riding a bike (or walking). And after the interview, they interviewed a bunch of their friends/family/etc. and came back to us with those results: Still the same, not a single story of an accident.
And this all goes back to the way the Dutch design infrastructure. A ton has been written/filmed/etc about this, but here's an example: https://www.youtube.com/watch?v=XuBdf9jYj7o
A dashed blue edge to the road means a lane separated by a kerb from cars and pedestrians. A solid blue edge means a line of paint on the road. There are also off-road paths with dashed, blue lines.
The thick, translucent lines are just signed, long-distance routes; they are likely to be mostly good, but not necessarily.
https://osm.org/go/0H54ISC?layers=C (Odense, a town in Denmark, just to show it isn't only Copenhagen.)
This product feels like attempting to put a technological band-aid over a poor infrastructure planning problem.
> Of course you can.
The road has to be wide enough to accommodate a bike lane. The absolute minimum width for a bike lane is 4 feet (not including the gutter adjoining the curb).
In my town, they recently restriped a 4 lane road with a concrete raised median and sidewalks on both sides to include bike lanes. They basically narrowed the 2 general purpose traffic lanes from 12 feet to 10 feet in width and added a 4 foot wide bike lane.
With the bike lane, if my bike with a 3 foot wide trailer is centered in the bike lane, I have about 0.5 feet to each side of me. A car with a mirror span of 7.5 feet centered in the 10 foot wide general purpose lane has about 1.25 feet on each side. 1.25 + 0.5 = 1.75 feet of clearance between a car in the right lane and a cyclist in the bike lane.
With the old configuration, a cyclist centered in the 12 foot wide general purpose traffic lane would have 4.5 feet of space on each side. A car centered in the lane would have 2.25 feet of space on each side. 4.5 + 2.25 = 6.75 feet of space between the car in the left lane and a cyclist in the right lane.
In that case, having a bike lane substantially decreased the possible passing distance between cars and bicycles. If they wanted to add the bike lane, then the road should have been at least 2 to 3 feet wider than it is now.
> every road here has a dedicated bike lane, and usually it's a separate path.
Those lead to issues at intersections where turning vehicles or vehicles entering the road may not see you in time when you're crossing.
Separated paths improve safety, they don't cause issues at intersections that I've noticed (and I've been biking through them every days for years), probably because things were designed well to account for this. It's one of those situations where good design becomes mostly invisible so it's hard to spot.
Not really. This document describes the crashes involving cyclists in a bidirectional separated bike path[1] which for a couple of years was having close to one crash per month.
Here's an example[2] of a drive-out collision.
Here's an example[3] of a right-hook near miss.
Here's an example[4] of a left-cross collision.
[1] https://drive.google.com/file/d/1o9UFZiSnBUOYsCBeY-nhnxboQgF...
[2] https://www.youtube.com/watch?v=4k6-AI_X1qE
[3] https://www.youtube.com/watch?v=-Yea-yFRGHg
[4] https://vimeo.com/32887898
Compare to here, https://goo.gl/maps/F1mufzKACT3XpDZa6, just as a fairly random example. You see that the car traffic is forced, via traffic calming (i.e. the raised footpath) to slow, has obvious lower priority than bikes and pedestrians, and can't really proceed without being strongly encouraged to look first. A couple more interesting samples at the bottom that are designed with safety in mind, and in my experience work well.
Taking an existing road, slapping a bike line on one side, and calling it a useful test is not really effective. This is why I mentioned design is important.
What is interesting from your PDF (which I just skimmed, so don't know if it's representative) is that the number of people cycling increased, so it looks like people felt it was safer even if it really wasn't. I suspect that if you managed to get that up to real numbers of people cycling, then drivers would naturally become a lot more aware also.
* https://goo.gl/maps/DLHUVjadJcM2y8tG7 - light controlled, bikes tend to sit slightly forward of the cars in order to be visible.
* https://goo.gl/maps/bhLYgqiVBccnq2i27 - roundabout with (in order of priority) trams, bikes, busses/cars
I also don’t see how this would do the “prevent accidents” part of “could detect when such collisions were going to occur to take action and prevent accidents”.
Who’s going to take action here? If it’s the cyclist, how are they going to tell them what to do, and how are the cyclists going to do the right thing, given, likely, less than a second to act? (If you’re going to warn them earlier, I would hope there will be lots of false alarms for cars that see you, but don’t take action yet. Or am I mistaken in understanding what kind of war zone cyclists live in in the USA? We aren’t talking about cyclists being pumped up with adrenaline all the time, are we?)
I would guess a rear-view camera or, way simpler, rear-view mirror would be the better and easier way to increase situational awareness of cyclists?
In environment where there frequent intersections, these actually lead to more problems for cyclists at those intersections due to risks of collisions from:
1. right-hooks where traffic makes a right turn and crosses the path of the cyclist
2. left-crosses where left turning traffic crosses the path of the cyclist
3. drive-outs where traffic pulling from a side-street or driveway crosses the path of a cyclist.
These are much less of an issue if the cyclist rides in the center of the right-most general purpose lane since motorists will be looking for traffic in those locations.
The hardware/firmware/software/AI necessary to make such a device wasn't available when we started this though, so we had to make it first. And we're releasing it with OpenCV. See here for details: https://opencv.org/opencv-spatial-ai-competition/
So the bike-safety idea is not a product yet, but the underlying platform is now (and is going live on KickStarter soon).
Thoughts?
Is this a product that end users (car owners) could buy? How would it be integrated with the car? As a driver who doesn't care about cyclists (as those are the ones causing most accidents), would you buy a specific device just for avoiding cyclists?
Is it a product car manufacturers have to integrate in the car? Shouldn't this have a wider target than cyclists? How does it compare to the existing collision avoidance systems? If it's an optional safety system just for detecting cyclists, how do you sell it to the car buyer?
https://discuss.luxonis.com/d/8-it-works-working-prototype-o...
> So the final product is a multi-camera, rear-facing bike-light, which uses object detection and depth mapping to track the cars behind you
It's for the cyclist.
I'm both a car driver and a cyclist. I wonder if I want something like this on my bike and be unnecessarily scared by false positives (cars do get close to bicycles all the times and what can I do anyway?) or on my car just in case I don't notice a bicycle with no lights at night (plenty of them here.)
Right?
Also a cyclist, I can totally imagine this going (uselessly) wild when I do something like lane-slip through peak-hour traffic in the city.
I've had situations where I rode past someone (they were stopped in traffic), then they passed me again down the road (this repeated a couple of times) and every time I rolled past them, they'd descend further into apoplectic rage, I can't imagine flashing lights is going to help.
When it happens to me to be a car driver I think that at least I'm enjoying heating or air conditioning but I'd like to be on my bicycle at least until I get out of the city.
Would like to give the benefit of the doubt here to the CEO and Intel, but being weary of Intel’s past “hacker friendly” blowharding and subsequent total abandonment of the hacker community with the Edison, I do wonder whether Intel is again playing the mafioso; the wording of this quote from the Luxonis CEO comes off, to me at least, as kissing the ring,
> “DepthAI, powered by Intel Movidius Myriad technology, is making it tractable.”
Also wonder about the nature of the open source software the article refers to,
> The Luxonis CEO said that the SoM works with a wealth of code that the company has open-sourced through an MIT license.
Is the software running on the Myriad X itself open source, or just the SDK interfacing with it from the host machine? If the former, pretty amazing they were able to get Intel to agree to this, as I imagine the chip details are under strict NDA. If the latter, it seems pretty disingenuous to refer to the SOM software as “open source,” at least not with a footnote that the most interesting part (the spatial depth calculation) is not, presuming that’s done on the Myriad X.
So although I could see how that quote could appear like that, the quote is not kissing the ring. We scoured the whole semiconductor start up scene to find a chip that could be used for this.
There are only 2 chips (as I'm aware, as of this writing) that can be used in such a way: 1. Intel Movidius Myriad X 2. Inuitive NU4000AI.
And the Inuitive, until super recently (i.e. 1.5 years after we have already built hardware off the Myriad X) was not ready to be used (had tape-out issues at the fab).
So this made the Myriad X the only chip that (1) had the capabilities needed and (2) was actually available and in production.
In terms of our open-source nature, the latter is what we've implemented - where we have closed source binaries running on the Myriad X - which then have a slew of open-source counterparts on the host. Sorry if this came off disingenuous. Do you have advice on how to phrase it in a better way?
https://github.com/luxonis/depthai-hardware - DepthAI hardware designs themselves. https://github.com/luxonis/depthai - Python Interface and Examples https://github.com/luxonis/depthai-api - C++ Core and C++ API https://github.com/luxonis/depthai-ml-training - Online AI/ML training leveraging Google Colab (so it’s free) https://github.com/luxonis/depthai-experiments - Experiments showing how to use DepthAI.
The above includes open source hardware, core capabilities in C++ which can be cross-compiled for various hosts, and open-source training notebooks and use-case examples.
So the goal of open-sourcing all that we can (including hardware) is to enable folks who have their own applications to leverage this, modify it, use it, etc. w/out having to even talk to us. We cannot open-source the code that runs on the Myriad X however, as then we wouldn't have a way to monetize, and would have to give up on the mission. (That, and we're also not allowed to.)
Thoughts?
Thanks, Brandon
What does it do with this information?
It's normally blinking/patterning like a normal bike light, and if it detects danger, it does one of two actions, and sometimes both:
1. Starts flashing an ultra-bright strobe - to try to get the driver's attention
2. Initiates haptic feedback - to let the person riding the bike that they're in danger
3. Honks a car horn. This is if all else fails, and only happens if you are in certain danger"
Source: https://discuss.luxonis.com/d/8-it-works-working-prototype-o...
The Radar powered Garmin Varia (https://buy.garmin.com/en-GB/GB/p/601468) is probably simpler and cheaper and I'd guess more likely to work reliably in bad lighting conditions.
But a good set of continuously flashing ultra bright strobe lights, high visibility/reflective clothing, and bike lanes or driver education is by far superior.
It's not intended for use in commuting though... as in commuting you don't care about the 500 cars that will pass you safely on your ride (and you don't want 500 alerts), you care about the 1 that is on a trajectory to kill you.
So with computer vision you can tell that trajectory. You can also with radar, but only with super-expensive radars ($100k+), wheras you can do this w/ computer vision for $50 or so.
edit: Still, I'd be afraid that falls from startled reactions by the cyclists themselves might in the end cause more bloodshed than the extremely rare rear-ender it's supposed to prevent. A car preparing a well timed overtake (close in fast inside the lane to sweep left just after oncoming traffic passes) will be a false positive you can't avoid without directly guessing intention and skill of the driver. It will surely be a net-positive in some communities with particularly low cyclist awareness amongst drivers but not necessarily in others.
So at the outset of this effort (back in 2017) such a computer-vision based device that was inexpensive and embeddable was impossible. The Movidius Myriad X came out, which has all the requisite CV/AI processing to make such a thing solvable... but there was no hardware/firmware/software/AI ecosystem which allowed producing such embedded spatial AI problems.
So we had to build that first to solve this safety problem and we're releasing it in conjunction with OpenCV as the OpenCV AI Kit (OAK). https://opencv.org/opencv-spatial-ai-competition/
So the `why` of this is it allows others to solve problems on embedded systems which were previously intractable. And there are a ton of them... the number of applications we've seen has blown our mind.
We'll have a KickStarter around this OpenCV AI Kit which will be going live on July 14th: https://www.kickstarter.com/projects/opencv/opencv-ai-kit
Thoughts?
Thanks, Brandon
It works really well, but requires the cyclist to assess the risk and react. Something like this, that could inform both myself and the driver of an impending collision, would be (literally) a life saver.
https://en.m.wikipedia.org/wiki/Vehicle-to-everything
Thanks for the reassuring words sunpazed! Yes, that’s exactly right... where the Garmin product is great for alerting the rider there exists a vehicle behind them - which is SUPER useful for touring riders - this alerts the rider and the driver of an impending collision. Which is what makes it applicable to bicycle commuting, where say 500 safe passes occur per commute, and you only care about the one that’s going to hit you.
Radar works great for presence, but is extremely expensive compared to computer vision for determining exact positioning and 3D trajectory. And computer vision (specifically the combination of stereo CV/AI) is actually super-good at this, and can see quite granularly (i.e. 1 million depth points every 60 times a second out to ~20 meters).
So unlink the Tesla example, where the driver knows his vehicle has this safety feature... the driver does not know the biker has this feature.
Thoughts?
My experience of cycling in the UK is many cyclists already go to great lengths to make themselves visible, and it makes no difference at all to close passing, etc. Hi-viz clothing, bright strobing lights - drivers act like you're not there.
It also makes drivers lazy - having nearly been knocked off my bike, the driver stopped and berated me for not wearing hi-viz, despite the bright lights I was using. Not impossible, therefore, that I also get berated for not using your device to warn me about drivers. There have been attempts to do this in the UK already with apps which notify drivers of cycists, with the app maker saying it would be the cyclists' fault if they were in an accident and weren't themselves using the app. (This was repudiated by the Advertising Standards Authority, but still.) This is similar to the all-too-frequent reporting about "A cyclist was crushed to death by a truck. Police said the victim was not wearing a helmet." Even the helmet manufacturers don't say it will stop a 20-ton vehicle from killing you :(
As Chris Boardman says about the helmet debate, it's the wrong emphasis. The issue is safety on our roads and drivers being careful around cyclists. Also, cyclists being predictable and riding safely. An electronic device will make little difference without that, and I feel it could make things worse.
Also, I find a mirror on the handlebars, a life-saver check when turning, and listening for engine noise to be pretty good at alerting me to drivers. Not sure electronics would improve this.
Protected bike lanes still need to cross regular streets (and that's where most accidents happen).
Technology won't solve people problem. Drivers need to expect bikers like they expect other cars and adjust speed to that expectation.
That's what we harp as well (Luxonis-Brandon here). And we actually see our goal as means to get there... to do what we can (as technologists) to try to keep people who ride bikes safe, keep people riding bikes, and keep there being demand for the infrastructure going.
If everyone is too afraid to ride bikes like my circle is (reminder: 1 of my friend was killed, and 3x got broken femurs, broken hips, and 1 of those got a traumatic brain injury), then there won't be demand for such infrastructure.
And with lower demand, it's harder for there to be incentive to make the infrastructure, so the infrastructure may actually get worse - making it even less safe - and then resulting in even fewer people riding bikes.
And there's another constraint: I'm an Electrical Engineer and such tech solution are all I’m good at. I don't know how to make roads with bike lanes. And more importantly, I don't know how to influence politicians to make these roads. (And my wiring is such that I'd never be good at it, no matter how hard I try.)
That, and there are ~4 million miles of road in the US. So if we put the whole economy onto making new roads as fast as we could, we'd be a decade or more realistically probably two or more decades out until we had the new roads with dedicated bike lanes.
This is what The Netherlands did, by the way, and they accomplished it (in a comparatively small country) in about 2 decades. And this accomplishment is why there exists absolutely zero car-bike accidents in the Netherlands (and much lower car-car accidents too).
So our goal is to try to help on the demand side. If there isn't demand for safe infrastructure, that infrastructure won't come. And there won't be demand if people are too afraid to ride bikes because of deaths and injuries of their friends/family.
So we want to try to use technology as a bandaid until the infrastructure is there, to try to keep people safe, and to also then help drive making infrastructure (by making demand for the infrastructure).
Speaking of which, this device quantizes risk. So when used on a bike, it would allow mapping quantized risk to bikers, which then helps to advise which are the most important routes/streets/etc. to be improved for safety. And also advises people who want to ride bikes as to which routes they should avoid.
So then when infrastructure is installed, this system can make it so the most-fatal locations get infrastructure first. So it's more prioritized and effective.
Thoughts?
Thanks, Luxonis-Brandon
And the core of the development is the computer vision technology which allows a whole slew of other applications, including other safety solutions. The sheet numbers of problems solvable with this sort of application-specific human level perception is crazy.
So we are actually doing a competition now around it, here: https://opencv.org/opencv-spatial-ai-competition/
We'll be releasing the winners of stage 1 soon (please feel free to apply!), which will quantify how crazy it is to be able to approach human-level perception (what an object is, and where) for specific myopic tasks (like picking strawberries, or detecting when a car is going to run you over).
Anyway, back to it, I agree personally the solution is infrastructure (which I've tried to write in every location that covers this... but maybe failed to do in that interview). But I view this as a stepping stone to infrastructure. If people are too scared to ride bikes because of the injury and death at the hands of distracted people driving cars, then there will be no push for infrastructure (as why build infrastructure for a thing no one does?).
So the goal is to allow people to ride bikes -safer- before the infrastructure is there... and thereby help drive the infrastructure change (through supply and demand relationship).
Thoughts?
Thanks, Brandon