Intel and Mobileye come up with a formula to verify the safety of self-driving cars

　　Professor Amnon Shashua , CEO of Mobileye and senior vice president of Intel, delivered a keynote speech at the World Knowledge Forum held in Seoul, South Korea, where he provided the autonomous driving industry with a way to prove the safety of autonomous vehicles. This solution was published in Professor Amnon Shashua's academic paper and popular summary paper, which proposed a formal mathematical formula to verify that autonomous vehicles can drive in a responsible manner and will not cause traffic accidents due to the fault of the autonomous vehicle. Let's follow the automotive electronics editor to learn about the relevant content.

　　Intel subsidiary Mobileye is a leader in autonomous technology and the world's largest supplier of advanced driver assistance systems (ADAS) cameras. With extensive experience in automotive automation and the evolution from ADAS to full automation, Professor Shashua and his colleagues developed a mathematical formula that can determine the responsible party and the party at fault in an accident when there is no human driver.

　　Their proposed responsibility-sensitive safety model provides concrete, measurable parameters for the human concept of responsibility, while defining a “safe state” in which an autonomous vehicle cannot cause an accident, regardless of how other vehicles react.

　　In his speech, Professor Shashua called on industry and policymakers to “work together to develop a standard to clarify accident liability” for use when human and autonomous vehicles inevitably come into conflict. He explained that all current regulations are built around the idea of ​​human drivers controlling the car, while autonomous vehicles require new parameters.

　　Professor Shashua said:

　　The ability to identify fault is important. Just like the most skilled driver in the world, a self-driving car will not be immune to accidents caused by actions beyond its control. However, the most responsible, observant, and careful drivers are less likely to cause accidents due to human error, especially if they have the same 360-degree vision and lightning-fast reaction time as a self-driving car.

　　Based on clear definitions of responsibilities agreed upon by the industry and regulators, the RSS model ensures that autonomous vehicles only operate and drive within a framework defined as “safe”.

　　Sam Abuelsamid, senior research analyst at Navigant Research who is involved in the Traffic Efficiency project, said:

　　Regulators and policymakers around the world are working to manage the deployment of autonomous vehicles while encouraging innovation, and a common, open way to evaluate the efficacy of technology seems to be a good start. Mobileye 's Responsibility Sensitive Safety Model is a good way to start a conversation. At least as an evaluation method, it does not limit anyone to a specific technology, while providing a good framework for the decision-making process within the control system.

    The above is an introduction to the formula proposed by Intel and Mobileye to verify the safety of self-driving cars in automotive electronics. If you want to know more related information, please pay more attention to eeworld. eeworld Electronic Engineering will provide you with more complete, detailed and updated information.