Intel builds drone system to avoid accidents

According to Bloomberg News, Intel has developed a system that will ensure that self-driving cars will not cause accidents when mistakes are made, so as to appease the public who is skeptical of self-driving cars and contribute to the acceleration of this type of car. Deployment on the road. The world's largest chip maker is releasing a set of standards based on mathematical formulas to control the behavior of robotic cars and trucks. Intel believes that if they are adopted, they will help clarify the responsibilities and faults of both parties in the event of a traffic accident. According to Dan Galves, Intel’s vice president of Mobileye, “Any self-driving car that is put into practical use will inevitably be involved in an accident. One thing is clear: that is, compared to human accidents, public vs. machines. The degree of tolerance that caused the accident was much less.” Mobileye, a maker of automotive technology, was acquired earlier this year by Intel. Some auto parts makers believe that cars increasingly need computing power, which is caused by the transition to autonomous driving, which is a new growth market. Intel also holds the same view. Although automakers, their suppliers, and companies such as Uber Technologies Inc. and Alphabet Inc. are testing on self-driving cars, Intel and its competitors need the industry to quickly pass the pilot phase and put it into production to make them Funds invested in R&D are rewarded. Intel is trying to establish a framework to prevent the chaos that can be caused when a self-driving car and those cars driven by unpredictable humans hit the road at the same time, and this situation is the only way to the future. The company has analyzed and analyzed the behavior and environment involved in almost all the accidents recorded by the National Highway Traffic Safety Administration (NHTSA) of the United States, and proposed a mathematical model to create a quantifiable "safety state" for self-driving cars. ". If it is recognized by the automotive industry, suppliers and regulatory agencies, these standards will also be the basis for the various software that the vehicles complying with the rules will carry. This will help speed the validation of automated driving technology and will also benefit Intel’s chip manufacturing business. According to Intel’s thinking, under this guideline, a robot car will drive past a parked car at a slow enough speed to ensure that it can stop in time to avoid pedestrians suddenly walking onto the road. Intel stated that this calculation is possible because we know the maximum speed one can move and can model it. Similarly, the computer can easily calculate the safe parking distance with respect to the vehicles in front and ensure that the vehicles they are driving have enough distance to stop. If an aggressive human driver steps in front of a robotic car and causes an accident, the standard will clearly show who is at fault, even if the robot car is following another car. Intel believes that the current path taken by the auto-pilot car industry is not feasible and takes too long. Slow-moving, over-cautious cars are limited in their use and unsafe because they block roads and are not suitable for the flow of traffic created by humans driving cars. Intel said that in order to prove that self-driving cars are safe, deploying them on the road, letting them learn from experience, and then measuring how much fewer accidents they cause are compared to humans driving cars, this approach is not feasible, in part. Any accident caused by a robot car has attracted widespread public attention.