Live data from Hacker News

BYD's cheapest electric cars to have Lidar self-driving tech

thedriven.io

91–100 of 340 posts

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#91
post #86
post #76

Earlier quoted context omitted.

Why is it preferable to wait for people to die and then sue the company instead of banning it in the first place?

They don't have to die first. The company can avoid the expense by planning how not to kill people. If you charged car makers $20m per pedestrian killed by their cars regardless of fault you'd probably see much safer designs.

> They don't have to die first. The company can avoid the expense by planning how not to kill people.

This is an extremely optimistic view on how companies work

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#92
post #76

Earlier quoted context omitted.

Nah, you don't need to ban anything. Just force the rule, that if company sells self driving, they are also taking full liability for any damages of this system.

Why is it preferable to wait for people to die and then sue the company instead of banning it in the first place?

We cannot even properly ban asbestos, expecting people to die first is just having a realistic perspective on how the US government works WRT regulations.

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#93

Earlier quoted context omitted.

Show the cost differences and do the math then come back to us before you can suggest what decisions were ill timed. Otherwise it's just armchair engineering.

I'd love to take on this challenge: the article they linked shows the cost add for LIDAR (+$130) -- -- but I'm not sure how to get data on ex. how much Tesla is charged for a Nvidia whatever or what compute Waymo has -- My personal take is Waymo uses cameras too so maybe we have to assume the worst case, +full cost of lidar / +$130

Camera's are not the issue, they are dirt cheap. Its the amount of progressing power to combine that output. You can put 360 degree camera's on your car like BYD does, and have Lidar. But you simply use the lidar for the heavy lifting, and use a more lighter model for basic image recognition like: lines on the road/speed plates/etc ...

The problem with Tesla is, that they need to combine the outputs of those camera's into a 3d view, what takes a LOT more processing power to judge distances. As in needing more heavy models > more GPU power, more memory needed etc. And still has issues like a low handing sun + white truck = lets ram into that because we do not see it.

And the more edge cases you try to filter out with cameras only setups, the more your GPU power needs increase! As a programmer, you can make something darn efficient but its those edge cases that can really hurt your programs efficiency. And its not uncommon to get 5 to 10x performance drops, ... Now imagine that with LLM image recognition models.

Tesla's camera only approach works great ... under ideal situations. The issue is those edge cases and not ideal situations. Lidar deals with a ton of edge cases and removes a lot of the progressing needed for ideal situations.

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#94
post #16

Keep in mind, that $25k AUD is just $16600. And for that price, you're getting a real car with driver-assist features and a reasonable crash safety rating. The US car manufacturers are cooked.

I think it's a little early to make that claim. Jim Farley is definitely paying attention, for example. He drove a Chinese EV for a year, IIRC, and on many occasions talked bout the challenges of competing with them.

I don't know what the real barrier to success will be, but I don't think it will be blindness. It may be difficulty competing on labor cost, but that's a good case for carefully applied tariffs to keep competition fair.

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#95
post #60

Earlier quoted context omitted.

This can't last indefinitely. At some point, the contrast between the US-made and Chinese-made cars will become too great to ignore. We saw that during the 80-s, with the Japanese cars.

Don't Americans like big pickup trucks? Nobody else really drives those in large numbers.

Yes we do. We have nice big wide roads. Heck, my European immigrant friends love trucks more than natives, in my experience. If you have the space for them, there are some very appealing attributes. My Lightning will carry anything I want, tow big trailers, has huge interior space for the family, will outrun most cars (even many 'fast' ones), and is more fuel efficient than a [non-plugin] Prius.

I wouldn't want to own it in a very dense city, but there are only a couple of those in the US. Most US cities even at their densest locations are fine with a half ton.

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#96
post #25

How do you train a model to drive with LiDAR when the human drivers who generate the training data don’t use LiDAR?

My impression was that the state of the are was still to generate high-level data from your inputs, then react with a mixture of ML and algorithmic rules to those inputs. For example you'd use a mix of LIDAR and vision to detect that there's a pedestrian, use past frames and ML to predict the pedestrian's next position, then algorithmically check whether your vehicle's path is likely to intersect with the pedestrian's path and take appropriate action if that's the case

Under that model, LIDAR training data is easy to generate. Create situations in a lab or take recordings from real drives, label them with the high-level information contained in them and train your models to extract it. Making use of that information is the next step but doesn't fundamentally change with your sensor choice, apart from the amount of information available at different speeds, distances and driving conditions

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#97
post #89

Earlier quoted context omitted.

The long-term view of LIDAR was not so much that it was expensive, though it was at the time. The issue is that it is susceptible to interference if everyone is using LIDAR for everything all the time and it is vulnerable to spoofing/jamming by bad actors. For better or worse, passive optical is much more robust against these types of risks. This doesn't matter much when LIDAR is relatively rare but that can't be ass…

I hadn't heard that criticism. You can get multiple Waymos near each other without them crashing into things.

When I see Waymos fail they usually fail together

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#99

Earlier quoted context omitted.

I'd love to take on this challenge: the article they linked shows the cost add for LIDAR (+$130) -- -- but I'm not sure how to get data on ex. how much Tesla is charged for a Nvidia whatever or what compute Waymo has -- My personal take is Waymo uses cameras too so maybe we have to assume the worst case, +full cost of lidar / +$130

The issue isn't just the cost of the lidar units off the shelf. You have to install the sensors on the car. Modifications like that at the scale that Waymo does them (they still have less than 10K cars) are not automated and probably cost almost as much as the price of the car itself. BYD is getting around this by including them in a mass produced car, so their cost per unit is closer to the $130 off the shelf price.…

Waymo already has an automated integration line, and the new vehicles from Zeekr will come partially assembled from the factory as a semi-custom design so there's no modifications in the sense that you're talking about.

Re: BYD's cheapest electric cars to have Lidar self-driving tech

#100
post #81
post #42

Earlier quoted context omitted.

What if the real world safety data over time is... secret retinal damage to millions of walkers and runners, with symptoms attributed to Covid mysteries (and not obviously due to vision), and it takes years more before someone happens to get enough data, and does the right study analysis, and then there's industry with strong incentive not to be on the hook for blinding millions of people? If the tech industry has ta…

Lidar is incredibly low power and fast scanning, the retinal risk is probably much less than having to drive when the sun is near the horizon.

However LIDAR safety is currently mostly evaluated on the assumption of a single LIDAR being present. If LIDAR becomes common, with multiple systems per vehicle, the probability of multiple LIDAR beams of different LIDARs hitting your eye at the same time goes up
Post reply on HN