Tesla is developing its own mapping and navigation architecture to address persistent issues with the Full Self-Driving (FSD) system. During the Q3 2026 earnings call, the company's VP of AI, Ashok Elluswamy, confirmed that Tesla will stop relying on external map providers to improve route planning, parking, and speed limit recognition.
The new system will utilize data collected from the existing Tesla fleet to build maps that adapt to driver preferences and support voice commands. The technology is scheduled to begin managing the start and end of trips in FSD v15, with navigation and parking capabilities expected to follow in the subsequent months.
This shift marks a departure from the company's previous stance, as CEO Elon Musk had long argued that relying on pre-built maps made self-driving systems brittle. Tesla now aims to use its fleet to map roads, a strategy that aligns with the operational methods of competitors like Waymo.
The move comes amid scrutiny regarding FSD performance. A recent study in Brussels showed Tesla vehicles exceeding speed limits in 30 km/h zones, while the NHTSA is considering an investigation into reports of Robotaxis speeding through school zones. Tesla has delayed the launch of unsupervised FSD for customer cars until Q2 2027.
