Toyota Motor is set to roll out an advanced self-driving system in its passenger vehicles from 2028, with wider adoption planned for 2030. This new system, labeled as "Level 2++," aims to provide significant driving assistance while still requiring human supervision, meaning the driver remains responsible for monitoring the road and intervening when necessary. Toyota's approach combines an end-to-end artificial intelligence for driving decisions with a separate, rule-based "guardrail" that can block unexpected or unsafe actions, addressing the "black box" problem often associated with AI-driven systems.
This Level 2++ system is designed to go beyond current adaptive cruise control and basic lane centering functionalities. It is intended to support more complex maneuvers such as lane changes, overtaking, navigation through junctions and intersections, and parking, even on urban roads. While Toyota will develop the underlying technology to a Level 4 capability, the commercial offering for passenger cars will be supervised Level 2++ to prioritize safety and ensure explainability. This strategy aligns with Toyota's long-standing "Guardian" concept, which emphasizes human oversight.
The core of Toyota's system involves an AI that analyzes sensor data holistically to manage steering, acceleration, and braking, rather than using compartmentalized modules. The unique "guardrail" mechanism will continuously monitor the AI's outputs against traffic regulations and safety thresholds. If the AI attempts a hazardous action, the guardrail will intervene, correct, or restrict the vehicle's movement, and log the intervention for post-incident analysis and liability clarification. This contrasts with some other autonomous driving systems, like Tesla's FSD, which primarily rely on camera-based vision and a single neural network, as Toyota's system will also incorporate lidar and radar sensors for enhanced obstacle detection and environment mapping.
