Autonomous Car Progress

[Knightshade]

1) Very inconvenient that I can't change lanes without aggressively disengaging AP.

Buy EAP or FSD. Both include automatic lane changes (both user initiated with a turn signal and car does the rest- and automatically initiated by car based on slower traffic)

Tesla is very unlikely to change existing basic "free" AP behavior in a way that discourages sales of the higher end versions of their ADAS systems.

 

[DanCar]

https://twitter.com/x/status/1639058097082859526

I wonder how the accident happened. Surely, the Cruise saw such a large object in front it. Seems odd that the Cruise would rear end it.

Cruise DMV report simply states they hit the minibus.

https://www.dmv.ca.gov/portal/file/cruise_032323-pdf/

Does this mean Cruise needs more lidars?

 

[DanCar]

Teens harass a Waymo. They run away when Waymo says to step away.

Car says: "Please step away."
Announcer states this about 1:50 into the video.

 

[Knightshade]

Teens harass a Waymo. They run away when Waymo says to step away.

Car says: "Please step away."
Announcer states this about 1:50 into the video.

It should've added "You have 30 seconds to comply"

 

[diplomat33]

Does this mean Cruise needs more lidars?

No.

 

[diplomat33]

Waymo had some fun for April Fool's Day.

Intro greeting to riders tells a joke:

And they put teddy bears in the driver seat:

https://twitter.com/x/status/1642136068160987138

https://twitter.com/x/status/1642067823899058176

 

[diplomat33]

Aurora says they are now L4 feature complete to begin their autonomous trucking between Dallas and Houston.

https://twitter.com/x/status/1642875356913995777

 

[ThomasD]

Since it won't have a backup driver if the Truck comes across a situation it can't handle do they control it remotely? If the Truck needs remote assistance does the remote driver require a CDL?

 

[diplomat33]

Since it won't have a backup driver if the Truck comes across a situation it can't handle do they control it remotely?

No. Remote assistance never controls the truck. Remote assistance will give the truck a hint and then the truck will use that hint to solve the situation on its own. That is how remote assistance works. So, the autonomous truck must be able to handle all safety cases on its own since it would not be safe to rely on remote assistance to prevent a crash. Remote assistance can only be used for non-safety issues where there is no risk of collision and there is time to give the truck a hint and let it figure things out.

If the Truck needs remote assistance does the remote driver require a CDL?

No since remote assistance is never controlling or driving the truck.

 

[Doggydogworld]

Since it won't have a backup driver if the Truck comes across a situation it can't handle do they control it remotely? If the Truck needs remote assistance does the remote driver require a CDL?

Aurora says they now have the "features" needed to operate w/o a safety driver, but I kinda doubt they will actually do it yet.

 

[KArnold]

Aurora says they now have the "features" needed to operate w/o a safety driver, but I kinda doubt they will actually do it yet.

Their projections say 2023. I assume that's for their planned expanding level 4 routes, which initially will be that Dallas - Houston corridor is my guess.

 

[rbt123]

Aurora says they now have the "features" needed to operate w/o a safety driver, but I kinda doubt they will actually do it yet.

Agreed. AFAIK there are still legal obstacles such as deploying the required flares and cones when a truck pulls over for 10 minutes.

 

[diplomat33]

Cruise is hiring for their scaling efforts in Houston. They seem to pay pretty well.

https://www.reddit.com/r/SelfDrivingCars/comments/12br9eo

 

[thewishmaster]

https://twitter.com/x/status/1643415738823368704

Wonder how often these stalls are happening in SF and elsewhere…

 

[diplomat33]

Cruise detailed how they have worked to improve their autonomous driving in inclement weather.

https://twitter.com/x/status/1643662771584921601

Some interesting points:

• Optical cameras: these sensors most closely approximate human vision, allowing us to distinguish between objects, colors, human direction and intention. While optical camera performance may degrade on rainy or foggy nights, our camera still performs reasonably well to pick up high contrast 2D detections from objects like headlights and stop signs.
• Long and short range lidar: gives us highly accurate point cloud 3D representations of scenes and objects surrounding an AV. While fog and rain lessen lidar range, we’ve found lidar to be a valuable tool to make localized measurements of fog density near each AV in real time.
• Radar: allows us to see longer range views and detect the speed of a vehicle or object. It’s especially effective at spotting vehicles on the road at a distance beyond what humans can even in heavy fog. Radar enables AV to recognize fast moving objects and their radial velocity. Radar is the least impacted sensor from weather conditions. Our in-house designed, high resolution radar system could easily detect moving objects up to 200 meters away.
• Audio: allows us to detect sounds near the vehicle, for example, emergency vehicle sirens.
• We’ve integrated many custom-designed sensor cleaners, wipers, and air puffers to minimize water droplet flare and condensation on sensor lenses across our suite.
• Early on in road data collection, certain weather-related scenarios were perceived as “phantom objects” in our vehicle’s path. These included: Large puddles which often reflected light from other objects, and “Rooster tail” splashes which trailed vehicles in motion in front of an AV. Strong reflections and headlight glare on puddles and wet roads also had the potential to trigger false positives. Understanding that these conditions could degrade vehicle performance, we proactively fed rainy road data into our robust perception and prediction stack, generating mountains of “splash and puddle data” in simulation to harden our fleet. As a result, our AI has learned to detect and filter out these phantom objects.
• To compensate for this, we redesigned how our planning and controls stack responds to more comfortably handle excessive slipping events. Just as a trained driver would, each AV now detects rain and adjusts its driving behavior. If it still experiences excessive tire slip, it responds by adjusting its plan in real-time, keeping our AV and its occupants on a safe trajectory.

Source: Inclement weather - A full stack challenge

 

[EVNow]

https://twitter.com/x/status/1643415738823368704

Wonder how often these stalls are happening in SF and elsewhere…

Lack of comments from Cruise and what they are doing to address the problem is equally troubling .... instead they put out something about how great Cruise is in bad weather.

 

[Dewg]

Lack of comments from Cruise and what they are doing to address the problem is equally troubling .... instead they put out something about how great Cruise is in bad weather.

Nothing released yet on why they hit the bus either.

 

[Supcom]

Lack of comments from Cruise and what they are doing to address the problem is equally troubling .... instead they put out something about how great Cruise is in bad weather.

If it starts raining, perhaps the car will take off again.

 

[diplomat33]

Lack of comments from Cruise and what they are doing to address the problem is equally troubling .... instead they put out something about how great Cruise is in bad weather.

Nothing released yet on why they hit the bus either.

Yeah, I find Cruise is not very transparent on safety at all. They did not release their safety data on 1M driverless miles like Waymo did. And they don't really respond when there are incidents or stalls.

 

[EVNow]

Yeah, I find Cruise is not very transparent on safety at all. They did not release their safety data on 1M driverless miles like Waymo did. And they don't really respond when there are incidents or stalls.

Having watched GM over the years .... I'm afraid Cruise has picked up GM's culture. Let us see how long they last .... won't be surprised if they are merged with GM.

I wish all these companies are IPOed like MobilEye. We'll see how far they'll last on their own.