During a high-stakes sensor test, a Tesla Model Y operating on Autopilot failed to avoid a fabricated wall, echoing Wile E Coyote cartoon mishaps. Filmed by engineering influencer Mark Rober, the incident raises urgent questions about Tesla’s camera-only approach, prompting experts to remain significantly skeptical.
In a camera versus lidar test, Tesla Autopilot careened into a Wile E. Coyote-style faux road barrier smack dab in the center of the road.
Although the majority of firms working on autonomous driving technology have used a combination of sensors, including cameras, lidar, radar, and ultrasonic, Tesla is adamant about exclusively employing cameras, Electrek shares.
The carmaker disabled radars that were already in place and eliminated them from future models.
The plan has not been successful so far because Tesla’s technologies are not even close to reaching level 2 of driver assistance.
Because its vision plus neural net system mimics a human driver’s actions and can adjust to different types of roads, Tesla will have an advantage over its rivals if it achieves autonomy, according to CEO Elon Musk.
Many have raised objections to Musk’s statements, citing the fact that he said Tesla would achieve “level 5 autonomy”—”under all conditions”—and the fact that lidar sensors overcome the limits of cameras in this regard.
That exact issue is illustrated in a new video by engineering YouTuber Mark Rober:
Rober puts a Tesla Model Y on Autopilot and puts it through its paces in a battery of tests against a lidar-equipped vehicle under varying environmental circumstances.
Even in moving, static-filled, light-blind conditions, the Tesla on Autopilot stopped for a child mannequin in the center of the road. However, in foggy or heavily rainy conditions, it was unable to stop:
Lidar’s superior detection capabilities in dense fog compared to camera systems are not unexpected, given that it is a laser-based device.
It was unexpected for there to be significant rain, but the amount of rain was really remarkable.
Despite the obvious impracticality of the fourth scenario—a Wile E. Coyote-style wall with a painted road—it does highlight the problem with cameras as opposed to radar or lidar sensors: they depend on human perception of possible barriers rather than objective data regarding such impediments.
To put it simply, lidar sensors were unconcerned with the artwork on the wall; what mattered to them was that it was a wall. In contrast, cameras are susceptible to manipulation.
Luis Gochoco is a seasoned managing editor and writer with over a decade of experience covering politics, technology, gaming, and entertainment news. With a keen eye for breaking stories and in-depth analysis, he has established himself as a trusted voice in digital journalism. Luis is one of the key forces behind the success of GameNGuide, contributing to 12 million views through engaging and high-traffic content. He also played a pivotal role in generating 8 million views on International Business Times, shaping the platform’s technology and gaming coverage.
Discover more from Today's Esquire
Subscribe to get the latest posts sent to your email.