stefan8r

Lidars break. Radars fail. Cameras run into issues. Drive by wire systems malfunction. Computers vibrate to death.

All parts of an autonomous system break. Safety engineering is measuring those breakages, and designing a system that is safe when it inevitably breaks.

If you're the operator, you can choose to only drive on routes that should have sufficient connectivity. If your remote driver is only issuing high level commands (i.e. not responsible for safety) latency starts to stop mattering so much.

The problem here is the availability bias - you've seen your phone fail so you know that telecom links can fail. As a layperson you might not think about how the rest of the system suffers from the same limitations - but they do. You engineer around them.

The difference between "in the sleeper cab" and "outside the truck" are huge. We first started doing in the sleeper cab in Jul'17, took us 2 years of dedication to get confidence to remove from vehicle.

To take the person out of the vehicle, the system itself needs to be able to self-monitor and pull itself over if it stops working. That's the big lift.

I haven't heard that thought before, but it's an interesting one.

Probably won't help with any of those folks opposed to UBI because "people get meaning from work." Proponents of that view must know a hidden trove of self-actualized retail workers.

TBH - In the early days the "cool AV people" seemed to act as if teleop was an admission of bad engineering (your team isn't good enough to perfect L5). If everyone else is right on the cusp, and you're not trying, you must be some sort of loser.

There are still a number of folks who think that teleop can never be safe - surprisingly sophisticated people hold onto that dogma. We wrote a whitepaper about it for investors, and I'll share that at some point as I open up more and more of the Starsky vault.

Answering these from easiest to hardest:

3) Sure! It isn't too hard to figure out how to email me, so send me a note or DM on twitter or LinkedIn or whatever.

2) AV has been a great testcase for federalism, and I think we've seen different regimes work differently. For the stage that industry is at, I think that insurance requirements is sufficient (insurance carriers have the sophistication to figure out if you're being unsafe). California's regime is overly-prescriptive in such a way that it confuses things more than it improves safety.

To be clear - if we were closer to deployment I'd advocate for a way tighter regulatory regime.

1) We figured out a programmatic way to measure ODD. Trying to specify all versions of ODD is a fools errand, and you're right that it would change a lot by how good you stack is.

ODD = Operational Design Domain. The conditions your system can drive safely in - rain, snow, sunshine, glare, high traffic, no traffic, pedestrians, etc ad infinitum

My current thesis is that robots (have to sense the world, make decisions, move freely, and respond to stimuli) only make sense in markets with labor shortages.

AKA - if people are willing to do it for cheap, it probably won't make sense to get a robot involved.

What's hard about this comparison is knowing exactly what you saw. If the safety driver disengaged 1-3x, it might have felt like a robotaxi drove you all over; but the effort to get that system safe for unmanned regular service might take another decade.

If it was super mapped and following a fixed route with object avoidance; the difference between a car that can go on an HD-map track and one that can go point-to-point in a city is also maybe a decade.

We humans have limited empathy for what is really really really hard for computers. Many know that's true for an app, but somehow ignore that knowledge when it comes to autonomy.

Some large teams have tried to overcome this by making up every such scenario, coding it into simulation, and testing their system.

What makes this challenging is that there are more such scenarios that might occur than there are grains of sand, or stars in the universe.

Those teams end up thinking they did all the necessary work and then go test, just to find out that there's a new edge case. At Starsky we called it Safety Theatre

Crisis PR is an interesting thing. Each incremental death from AV causes less repetitional damage. Uber had a big shitstorm when they first killed someone, now they kill someone every day and the response is crickets. This is less of an issue than people like to make of it.

Exactly. Tesla is an EV company that uses Autopilot to get better margins (and marketing)

There was a person in the vehicle. Amusingly - many VCs who "should know better" fell for the same error. I'd regularly get in arguments with investors about whether others had done unmanned or not.

Fair point. I'm shitty at drawing mathematical graphs.

I'll update at some point with different notation.

The problem is that most of those businesses don't make sense with the current cost of talent. The L2 systems OEMs are rolling out are primarily made by engineers in low-cost areas at $60-90k/yr, not $130-250k/yr in SV.

Realistically, most of the AV talent will go to industries that can afford them. Salary isn't super elastic.

There are a couple of people working on that, notably CloudTrucks.

It's not for the faint of heart - truck drivers and engineers follow a very different set of social norms. It's hard to be the hardass boss drivers don't want to fuck with while being the cuddly CEO engineers like the culture of. That might be a future blogpost.

All transportation businesses are utilization businesses. Empty hours/miles are never recovered.

Early robotaxis will have tight geofences. Makes them particularly bad competitors to poorly paid people who are willing to drive wherever.

Great point by the other responder.

High level - Uber and Lyft have done a terrific job getting people to work for less than minimum wage. I've met quite a number of uberdrivers who work in Robotaxi-Ready cities (think Chandler, Tampa, etc) who work 12-16 hours/day, 6-7 days a week, and make $20k/yr after depreciation.

While sensors are getting cheaper, most teams respond by throwing more sensors on the vehicle. It seems unlikely there will be a <$20k robotaxi anytime soon.

TL;DR: Humans are cheaper than robots.

Hard to say. Truck drivers will be hit hard by this - they can't really stop interacting with people, almost never get paid time off, are older and in poor health, and are paycheck to paycheck.

My bet is that nearly all of them get COVID, with a higher hospitalization / mortality rate. Could knock 30% of capacity off the road in the next 2 quarters, which would mean food shortages.

I live next to the weird intersection at Duboce Park. I see a Cruise car go by it every 30m.

A better solution for robotaxis - just avoid the weird intersections.

Another common fallacy. The US actually has a better rail network than Europe - most people just think it's worse because passenger rail here sucks. That's the case because we prioritize freight trains over passenger (in America Business is #1).

To paraphrase Churchill, trucking is the worst form of transportation, besides all the others. If you can put freight on a train, you do. It's slower, point-to-point, and less good a less-than-trainload freight than trucks are.

Many many smart people share this thought. I don't.

Freeway costs roughly $4m/lane/mile. Building a duplicate system for trucking would make UBI look cheap. Truck taxes also pay for much of current hway infrastructure, so you'd need to replace that $

Hard to say.

We had really strict rules on what we wanted our data to look like, and were very specific about subject matter. We probably wouldn't have been able to be a customer.

Our intention to deal with accidents/edge cases was that if anything looked outside of ODD (as in, not perfect driving conditions) we'd execute a MRC (pull to side of the road, or stop at next exit). Relatively easy way to solve most of the really hard edge cases.

Also happy to answer questions (my username might sound similar to the name of the author ;) )

Just stopping is ok is ok in some use cases, but not others. It will be a long time before robotaxis "just stop" less frequently than uberdrivers, but AV trucks can already be more reliable than contemporary drivers.

This is true, but essentially every transit system in the world (minus Hong Kong and Singapore's) operates at a loss. Maybe you can save money vis a vis unionized drivers, but I'm not wildly optimistic.

System worked. Problem was investors have mostly bet on full autonomy, and when that failed to materialized they got scared out of the space.

Full autonomy isn't necessary. And I don't know if it's even profitable for trucks.

Starsky CEO here. Not an unrealistic thought at all.

My childish/idealistic personal thesis is that if you think 10-20% more, you can make "doing the right thing" into "doing the long-term profitable thing." Most business leaders are just too lazy / short-sighted to put in the extra effort.

Part of why we're being the operator is to make sure the teleop drivers can be paid well. Now, we might have competitors in time who mess that up, but if regulators which I think can be prevented by regulators and other stakeholders

Starsky CEO here: surprisingly it's both. Truck drivers have limited productivity (freight rates set by the market, cant drive more than one truck at a time), but it's also a large fragmented industry where many players have a zero-sum mindset.

Shippers often take too long to load your truck, which costs you scarce driving time. Your carrier might stiff you on you last weeks pay, because driving is too hectic and they're too far away for you to chase them in small claims court.

There arent enough drivers, but the industry also has many bad actors who are mean to the drivers that there are.

Starsky CEO here: Currently, cargo theft is less of an issue than F&F would have you believe, and unfortunately seems more driver related than land piracy.

It certainly is something that I think of, but safety engineering requires simpler processes (because simpler means less things need to go right, which is good when things are breaking). I'd rather pay more for cargo insurance than have more complicated logic that causes more accidents with people.

Not that I know of: none of those folks have actually taken the person out of the vehicle, and getting someone in the truck is the main problem the industry faces.

Similar some of to the others - we've hauled freight autonomously (we've actually hauled quite a bit of it), but we've also done yard jockey work in truck yards and done whole trips that didn't require a person in the vehicle.

You DARPA Grand Challenge guys certainly built some impressive stuff back then, and it's hard to shake a stick at what those teams accomplished.

The problem is, most of the industry has spent the last decade adding neat features and trying to improve their reliability...without much practical thought as to increasing controllability (in terms of accident avoidance) in the inevitability of a component failure.

Controllability has basically been shortcut by keeping a driver behind the wheel. We've done pretty significant testing - regularly hauling freight on the highway while in self-driving, no-one behind the wheel in super unstructured environments, dealing with extreme edge cases w/o intervention.

The big technical accomplishment that we're showing here, is that our system is controllable in the inevitable case of failure...without using a physically-present human as a crutch.