Poor Tesla Summon and industry wide driverless vehicle performance shows why DoD simulation technology is needed
Setting aside Tesla’s reckless, self-defeating and ridiculous avoidance of LiDAR and detailed mapping, especially at this stage of the game, the performance of Summon is really bad. Having said that no one is much farther along. Why? First, it is clear machine learning, especially around perception, is severely limited. Hyper detailed detection and lack of broader inference is a problem. However, that cannot be improved much or even be successful once remedied if folks keep using public shadow and safety driving for most of the development and test. The approach is simply untenable from a time, money, safety and liability POV. (More on that in my articles below.) The larger problem is the solution, using simulation for most of that development and test, is not a good option because the simulation technology in this industry is woefully inadequate. Particularly from a real-time and model fidelity standpoint. Most simulation companies don’t even try to get half the model type physics right. Why? Because the core architecture s for both is the wrong one. How can that be remedied so a true and complete digital twin can be made? Use of DoD/aerospace simulation technology. All of the issues can be solved. No Hype.
(Regarding whether simulation can replace the real world. Of course, it can. And do a far, far, far better job. Please see my article below.)
Let’s look at Summon as an example.
If DoD simulation technology were used every single model type in those parking lots would have a visual and physical digital twin. The exact vehicle, tires, road surface, sensors, noise and clutter from other sensors, all moving and fixed objects, weather etc. That includes that exact parking lot features and moving objects like people and other vehicles. Down to 5cm or less positional fidelity and mm visual and physical fidelity if needed. With that fidelity you can find problems you cannot with poor or generic models. And you can repeat things exactly. Without this high level of fidelity, you will often wind up with false confidence and real world tragedies.
For more please find more here
Proposal for Successfully Creating an Autonomous Ground or Air Vehicle
Simulation can create a Complete Digital Twin of the Real World if DoD/Aerospace Technology is used
Using the Real World is better than Proper Simulation for Autonomous Vehicle Development — NONSENSE
Autonomous Vehicles Need to Have Accidents to Develop this Technology
The Hype of Geofencing for Autonomous Vehicles
· https://medium.com/@imispgh/the-hype-of-geofencing-for-autonomous-vehicles-bd964cb14d16
SAE Autonomous Vehicle Engineering Magazine-End Public Shadow Driving
· https://www.nxtbook.com/nxtbooks/sae/ave_201901/index.php
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.
