Tesla will begin delivering the Model Y and Model 3 without radar to customers in Europe and the Middle East, a continuation of the company’s plan to use only cameras combined with machine learning to power its advanced driver-assistance systems and other CEO Elon Musk. Expectations of Active Safety Features.
Tesla first started making the Model 3 and Model Y without radar in May 2021. These vehicles are for the North American market only. The company said in an update to its website that it will begin delivering the camera-only Model Y and Model 3 for Europe and the Middle East this month.
The company said the vehicles may be delivered with some features temporarily limited or unavailable, including the ability to keep them in their lanes. The feature, known as Autosteer, will be limited to speeds of up to 80mph and a longer minimum following distance.
Tesla will slowly restore these features through a series of OTAs over the next few weeks. All other available Autopilot and Full Self-Driving features will be activated upon delivery, depending on order configuration.
Tesla’s decision to stop using the sensor is a departure from industry standards. Automakers often use a combination of radar and cameras to provide advanced driver assistance system features. Some automakers such as Mercedes-Benz and Volvo have even added a third sensor, lidar, to provide redundancy in ADAS systems.
Moving toward more — not fewer — sensors is the goal of most automakers. Tesla and its CEO Elon Musk did the opposite. Musk has long touted the potential of his brand’s “Tesla Vision” system, which uses nothing but cameras and neural network processing to detect and understand what’s going on in the environment around the vehicle and then react appropriately.
Neural networks are a form of machine learning that works in a similar way to how humans learn. It is a complex form of AI algorithm that allows computers to learn by identifying patterns in data using a series of connected networks.
Many companies developing autonomous vehicle technology for robotic axles or self-driving trucks use deep neural networks to tackle specific problems. But they isolate the deep web and use rule-based algorithms to work with the wider system. These self-driving developers also use a combination of cameras, radar and lidar to provide redundancy for the system.