Elon Musk, the CEO of Tesla, has been vocal about his stance on the vision vs lidar debate for autonomous driving. In a recent statement, Musk not only reiterated his belief that sensors like lidar are unnecessary for achieving self-driving capabilities, but he also went a step further by stating that using lidar can actually make self-driving cars more dangerous.
The debate was further fueled by comments made by Uber CEO Dara Khosrowshahi, who expressed his support for Waymo’s approach to autonomous driving. Waymo’s method involves outfitting cars with equipment such as lidar and radar to achieve superhuman levels of safety. Khosrowshahi believes that a combination of camera, radar, and lidar is the right solution for self-driving cars, despite the higher cost of implementing lidar technology.
In response to Khosrowshahi’s comments, Musk took to Twitter to explain his position on the matter. He argued that the use of lidar and radar can actually decrease safety by creating sensor ambiguity. When these sensors disagree with the cameras, it raises the risk of accidents rather than reducing it. Musk highlighted Tesla’s decision to focus on a vision-only approach, which has led to improved safety in their autonomous systems.
Tesla’s success with their vision-based autonomous systems was evident when they launched a Robotaxi pilot in Austin and the Bay Area. The company’s Full Self-Driving (FSD) technology, which relies on vision-based systems, has also shown promising results and is inching closer to surpassing human driving capabilities.
Musk’s comments have reignited the debate on the use of lidar in autonomous vehicles, with Tesla’s success serving as a testament to the effectiveness of vision-based systems. As the race for autonomous driving technology continues, it remains to be seen whether lidar will remain a staple in the industry or if vision-based approaches will become the new standard.