Article link — https://scale.com/blog/is-elon-wrong-about-lidar
This is the first test I have seen where someone tested camera sensor/perception systems with LiDAR. While the test seems to point to cameras having severe shortcomings, I think the test missed an opportunity to be more conclusive. They should have re-enacted the five Tesla “Autopilot” death scenarios especially the one that killed Joshua Brown and Jeremy Banner several years apart. They should have also added tests with stationary objects like firetrucks, construction barrels and even pedestrians standing still or baby carriages.
My position on the issue is that the sensor system, regardless of what it is comprised of, needs to detect objects and degraded objects properly in all relevant conditions and have redundancy. If that is two camera systems with sound detection — fine. Right now though no sensor handles location and object detection nearly good enough to settle on one type. I also believe that Elon Musk as well as Nissan and Wayve are reckless for suggesting or stating LiDAR is not necessary. These folks should prove their statements have any validity, especially Elon Musk and Tesla who have killed five people in a Tesla using “Autopilot”.
Finally — where is NHTSA? The NTSB? It’s one thing for the press and public to buy the hype and not do their due diligence. It’s not OK for that to include NHTSA and the NTSB. It is nonsense for the government to defer to industry and not provide objective safety bars. No industry has EVER shown it can self-police. My next article will be on this. (To include the first children being harmed in an autonomous vehicle under development as Guinea pigs.)
Please find my articles below that address each of these myths. (As well as a relevant bio)
Using the Real World is better than Proper Simulation for Autonomous Vehicle Development — NONSENSE
NHTSA saved children from going to school in autonomous shuttles and leaves them in danger everywhere else
SAE Autonomous Vehicle Engineering Magazine-End Public Shadow Driving
The Hype of Geofencing for Autonomous Vehicles
Autonomous Levels 4 and 5 will never be reached without Simulation vs Public Shadow Driving for AI (This article has links to most of the data my POV is derived from)
My name is Michael DeKort — I am a former system engineer, engineering and program manager for Lockheed Martin. I worked in aircraft simulation, the software engineering manager for all of NORAD, the Aegis Weapon System, and on C4ISR for DHS.
Key Industry Participation
- Lead — SAE On-Road Autonomous Driving SAE Model and Simulation Task
- Member SAE ORAD Verification and Validation Task Force
- Member DIN/SAE International Alliance for Mobility Testing & Standardization (IAMTS) Sensor Simulation Specs
- Stakeholder for UL4600 — Creating AV Safety Guidelines
- Member of the IEEE Artificial Intelligence & Autonomous Systems Policy Committee (AI&ASPC)
- Presented the IEEE Barus Ethics Award for Post 9/11 Efforts
My company is Dactle
We are building an aerospace/DoD/FAA level D, full L4/5 simulation-based testing and AI system with an end-state scenario matrix to address several of the critical issues in the AV/OEM industry I mentioned in my articles below. This includes replacing 99.9% of public shadow and safety driving. As well as dealing with significant real-time, model fidelity and loading/scaling issues caused by using gaming engines and other architectures. (Issues Unity will confirm. We are now working together. We are also working with UAV companies). If not remedied these issues will lead to false confidence and performance differences between what the Plan believes will happen and what actually happens. If someone would like to see a demo or discuss this further please let me know.