A research team from Tsinghua University has developed a breakthrough technology in autonomous driving that allows vehicles to self-learn while driving in unfamiliar situations. This innovation has the potential to resolve safety concerns related to self-driving cars. Unlike conventional methods, which require extensive training in various driving scenarios in advance, this technology enables continuous performance improvements by collecting data during autonomous driving.
According to the study findings published in the journal Nature Machine Intelligence, the newly-proposed technology outperforms conventional self-driving technology, which is based on an algorithm with more data collected through lengthy trainings in possible driving scenarios. With the conventional method, the car has a preset response plan in case of an emergency. However, the car might not know how to react in unfamiliar situations in which it has not trained, posing a threat to driving safety.
The research team evaluated the new technology through simulations and road tests, which showed that cars could learn in new situations and continuously improve their performance as driving mileage and data volume increased. The technology was even applied in vehicles at the 2022 Beijing Winter Olympics, and the team plans to further verify it on open roads with multiple scenarios. This breakthrough could be a game-changer for the self-driving industry and pave the way for safer and more efficient autonomous vehicles in the future.
Read More: