Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Can level 5 Autonomy be achieved with Hardware suite 2.0?

This site may earn commission on affiliate links.
You realize there are four side cameras right... (and they overlap viewing angles)

There are only two side cameras, in both sides on B-pilars. See

A look at Tesla’s new Autopilot hardware suite: 8 cameras, 1 radar, ultrasonics & new supercomputer

Edit;ok, Tesla uses name side camera also from rearward looking side cameras, but my point is, that there are only one camera on each side which has visibility 90 degrees to the side and if that camera on the B pilar is blocked Tesla cant see at 90 degrees angle to the side.
 
Last edited:
  • Disagree
Reactions: MP3Mike
There are only two side cameras, in both sides on B-pilars. See

A look at Tesla’s new Autopilot hardware suite: 8 cameras, 1 radar, ultrasonics & new supercomputer

Edit;ok, Tesla uses name side camera also from rearward looking side cameras, but my point is, that there are only one camera on each side looking 90 degrees to the side.

No see the tesla website (or check the graphic in the link you just posted). There are two forward facing side cameras and two rear facing side cameras.

Three in the front, four on the sides, and one in the back equals eight cameras.

(the rear facing side cameras are in front of the forward facing side cameras for overlapping coverage)
side-repeaterautopilot.jpg

side-pillarautopilot.jpg


The forward facing camera faces forward. There are NO cameras facing directly to the side with a narrow viewing angle.

tesla-second-gen-autopilot-sensors-suite.png
 
Last edited:
.................
Last Thursday, due to a road accident ahead, I was required to do a U turn (which is normal in my state, but technically illegal) along a road outside of a hospital. there were 2 ambulances there, but the police had not arrived. I followed the direction of citizens traffic control (plain clothed nurses were directing the traffic)
could driverless cars handle this, No.
could autopilot car handle this, yes with driver temporarily taking over control of the vehicle.

I think the only possibility is that a remote driver will move the autonomous taxi to a location where it can resume its journey. More work for call centers in India.

If Google gets to the point of a beta taxi, the first step is probably to simply move the person currently sitting in the car to a remote networked location. A single google taxi would still be monitored full time but remotely. Then the second step would be for that engineer to monitor multiple google taxis.
 
Cameras which Tesla calls forward looking side cameras are in the B-pilars, right? And these cameras provide the visibility 90 degrees to the side (among other directions). These are the only cameras providing visibility at 90 degrees angle.
 
Cameras which Tesla calls forward looking side cameras are in the B-pilars, right? And these cameras provide the visibility 90 degrees to the side (among other directions). These are the only cameras providing visibility at 90 degrees angle.
true, and for redundancy you have ultrasonics to 8 meters.

Elon was mentioning that if a camera was obscured due to mud/dirt etc. The car would let you know.

If there's a bus on the side of you and it obscures the camera then that's fine... it'll see a bus and the ultrasonics will agree. Why would it need to see past a bus to the side in a perpendicular view to the direction in which you are travelling? Cars don't often go sideways.
 
Last edited:
If there's a bus on the side of you and it obscures the camera then that's fine... it'll see a bus and the ultrasonics will agree. Why would it need to see past a bus to the side in a perpendicular view to the direction in which you are travelling? Cars don't often go sideways.

Tesla is coming from that hallway. No matter if it wants to turn left or right, it needs to yield that Mercedes Van, which is coming behind that bus. So it needs to see the van and it is in 90 degrees angle when B-pilar camera is blocked by that bus so it can't see it.
 
I wouldn't argue against the utility of "whitelisting" regions where or conditions when the car is most likely to function properly. But we can't also refer to such a system of caveats as being "fully autonomous". The industry will wind up needing to qualify what "fully autonomous" actually means compared to the sensibilities of a human driver.

How about a scenario where the destination address happens to be a building on fire. Would the "fully autonomous" car find a nice spot next to a blaze or would the AP heat sensors detect the danger? Can the car recognize a road closing (police, crowd, event, etc) several blocks away and plan a detour? How can the car tell the difference between a cop in the middle of the road directing traffic and a cop telling it to go back the other way? For that matter, how can it even understand a cop directing traffic? How would it handle a live power line draped across a parallel parking spot?

While I think the hardware may be sufficient for "L5", I think it will always be L5 in quotation marks. These are the kinds of things where a human will remain superior to an autonomous car for many years to come. Will an autonomous car be able to do this by 2018? No. Ever? Maybe, but what if the company decides such scenarios don't happen often enough to bother programming for? There will be situations where the "L5" system will have to give up and cry for help. Get ready to call a tow truck or find some other transportation so you can go rescue your "L5" car.
 
Tesla is coming from that hallway. No matter if it wants to turn left or right, it needs to yield that Mercedes Van, which is coming behind that bus. So it needs to see the van and it is in 90 degrees angle when B-pilar camera is blocked by that bus so it can't see it.

I can't even tell if that hallway is a drivable road or not (as a human). As electracity mentioned as a human you'd have to pull out pretty far see around that corner... the difference between your head and the B pillar is what, a few inches?

In addition, if the radar can bounce under a car and register a car in front then it could bounce off the opposing walls.

Also 150 degrees in the wide forward camera is pretty good. This gives you only a 15 degree blind spot which means once the car pulls forward it can "see" down the road a pretty good ways. Ultrasonics can check for pedestrians.

If you pull out and you are looking 90 degrees then you're already in the middle of the road. ;)
 
There will be situations where the "L5" system will have to give up and cry for help. Get ready to call a tow truck or find some other transportation so you can go rescue your "L5" car.
Agreed that it would have to give up and be rescued, but the rescue location could easily be in a location that's blocking the flow of traffic causing a huge problem. In that case, maybe the car should at unlock doors, put on a special hazard light and let any person move it out of the way.

Anyway, people won't tolerate an unoccupied car unnecessarily blocking progress on a road or through a parking lot. One could easily see it getting vandalized as a result of "stupid" behavior.
 
I think the only possibility is that a remote driver will move the autonomous taxi to a location where it can resume its journey. More work for call centers in India.

If Google gets to the point of a beta taxi, the first step is probably to simply move the person currently sitting in the car to a remote networked location. A single google taxi would still be monitored full time but remotely. Then the second step would be for that engineer to monitor multiple google taxis.
bingo
In mining, this is today's reality
Robot lidar trucks, supervised by a call centre 1200 Km away Driverless trucks move iron ore at automated Rio Tinto mines

the expected fuel saving didn't eventuate, but (for FMG) the elimination of toilet breaks basically paid for the system.
 
My major question is: If Tesla determines that AP 2.0 hardware is inadequate for level 5 in the next year, will they adjust the hardware setup before the Model 3 is released or will those of us with pre-orders get screwed...
 
My major question is: If Tesla determines that AP 2.0 hardware is inadequate for level 5 in the next year, will they adjust the hardware setup before the Model 3 is released or will those of us with pre-orders get screwed...

The problem with AP1 is: no existing wires, holes for additional hardware and especially, you need to create plumbings to liquid cool your nVidia Drive PX 2 supercomputer.

Once you've gotten AP2, whether you have to trade in your cars or not depend on how extensive is an upgrade.

Minor:
A change of cameras, radar.
Adding/changing nVidia Drive PX 2 motherboards...

All wires and plumbings are there, so I consider this kind of upgrade is possible (as opposed to the policy of trade-ins for new refreshed.)

Major:
Adding on additional Radar, LIDAR. I don't think there are existing wirings for them, thus, you need to trade your car in.