Waymo rejects camera-only systems for fully driverless cars
Waymo has criticised the camera-only approach to autonomous driving associated with Tesla, arguing that cameras alone cannot safely support fully driverless vehicles at scale. The intervention comes as Tesla prepares to launch its steering wheel- and pedal-free Cybercab, sharpening a long-running disagreement over the sensors and maps robotaxis should use.
Waymo says its more than 200 million fully autonomous miles show the value of combining cameras with lidar and radar, providing redundancy if one system fails. Tesla maintains that cameras can deliver safe, affordable autonomy and has dismissed lidar as unnecessary, while most other autonomous-vehicle firms use multiple sensor types; proposed New Jersey rules requiring several sensors could therefore exclude Tesla. Waymo also favours high-definition maps, updated with AI, whereas Elon Musk argues vehicles should operate effectively without them because roads constantly change.
- Waymo says camera-only autonomy is insufficient for safe robotaxis.
- Tesla argues cameras alone can enable affordable self-driving vehicles.
- Sensor requirements could restrict Tesla in some robotaxi markets.
New here? Start with this
Waymo and Tesla are developing cars designed to drive without a person controlling them. Such vehicles use sensors and software to detect roads, other vehicles and hazards, but companies disagree over which technology is needed to do this safely and affordably.
Waymo combines cameras with lidar, which measures distances using laser light, and radar, which uses radio waves to detect objects. Tesla relies on cameras and argues they can provide enough information; the companies also differ over whether cars should use detailed digital maps.
The debate matters as Tesla prepares to introduce the Cybercab, which has no steering wheel or pedals, and governments consider rules for driverless vehicles. Requirements for particular sensors could affect which companies are allowed to operate, while the choice of technology may shape the cost and capabilities of future robotaxis.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
A camera-only system could make driverless cars simpler and cheaper to build, which may help bring autonomous transport to more people. Supporters argue that cameras capture the visual information human drivers use, and that vehicles should learn to handle changing roads without relying on costly, frequently updated high-definition maps.
The case against
Fully driverless vehicles must detect hazards reliably across varied conditions, and relying on cameras alone may leave them vulnerable when visibility is poor or a sensor fails. Supporters of multiple sensors argue that lidar and radar provide complementary information and redundancy, while detailed maps can help vehicles anticipate road layouts; the safety stakes justify that added complexity and cost.
AI Americas Celebrity Cybersecurity Entertainment Software Technology World
Read the full article at the source →
Originally published by The Verge as “In a swipe at Tesla, Waymo says ‘cameras… aren’t enough’”.