SAE Level 4 Complexity Formula Leveraging Baseball and Richter Scale Analogies
I have been thinking for a while on how to communicate the very nonlinear progression of trying to reach SAE L4. What I came up with leverages four bases in baseball and the base-10 logarithmic Richter scale used to quantify the severity of earthquakes. With each whole number level being 10 times greater than the last. If that is extrapolated getting to home plate is 1000 times harder that getting to first base. Extrapolated more precisely, first base will take 1 billion equivalent miles of development and testing and getting to home plate would take 1000 billion or a trillion miles. Given this is the number Gil Pratt from Toyota gave years ago and is double the extrapolated RAND calculation of 500 billion miles. (RAND’s estimated 11 billion miles to be 20% better than a human. If one assumes an end-state of 10X better than a human, the final number is about 500 billion miles.) Keep in mind this does not factor in the thousands of human Guinea pigs injured and killed inside and outside the vehicles. And this is by company.
Now here’s the rub. These miles calculations ASSUME general and deep learning, simulation technology and the overall development and testing approach is tenable from a time, cost, and safety POV. Since NONE of these are remotely close right now, no one will get close to home plate. For ANY significant urban environment or highway. (It also assumes a competent sensor system. Which rules Tesla and Wayve out)
This now leaves us with defining the bases. (Assuming none of the issues I just raised)
· First Base — 1 billion miles = Micro ODD — Location, weather, scenario complexity and object differentiation and quantity. Basically, Waymo in Chandler Arizona with its grid pattern streets, good weather and with low complexity scenarios. To include all relevant crash, edge and long-tail cases.
· Second Base — 10 billion miles = Same location only now factor in weather and object complexity separately. To include all relevant crash, edge and long-tail cases.
· Third Base — 100 billion miles = Same location, now factor in weather and maximum complexity combined. To include all relevant crash, edge and long-tail cases
· Home Plate — 1 trillion miles = Full L4 at 10X a human’s average capabilities for all roads in the given country. To include all relevant crash, edge and long-tail cases.
I would like people’s opinions on this. I would think the most discussion would be on second and third base. But I could be wrong.
Below are a couple articles that explain my POV in more detail.
The Autonomous Vehicle Industry can be Saved by doing the Opposite of what is being done now to create this technology
How the failed Iranian hostage rescue in 1980 can save the Autonomous Vehicle industry
Waymo and Cruise should prove their systems are legitimately L4
The Most Dangerous, Deceitful and Deadly days in the Autonomous Vehicle Industry are upon us
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.
Industry Participation — Air and Ground
- Founder SAE On-Road Autonomous Driving Simulation Task Force
- Member SAE ORAD Verification and Validation Task Force
- Member UNECE WP.29 SG2 Virtual Testing
- Stakeholder USDOT VOICES (Virtual Open Innovation Collaborative Environment for Safety)
- Member SAE G-34 / EUROCAE WG-114 Artificial Intelligence in Aviation
- Member CIVATAglobal — Civic Air Transport Association
- Stakeholder for UL4600 — Creating AV Safety Guidelines
- Member of the IEEE Artificial Intelligence & Autonomous Systems Policy Committee
- Presented the IEEE Barus Ethics Award for Post 9/11 DoD/DHS Whistleblowing Efforts