Autonomous driving
Vehicle navigation without continuous human steering, combining perception, mapping, planning, and control, often using machine learning on sensor data.
Autonomous driving refers to vehicles that sense their environment, plan a route, and control steering, throttle, and braking without continuous human input. Practical systems fuse data from cameras, radar, LIDAR, and inertial sensors to build a map of obstacles, lane markings, and road geometry. Machine learning models often classify objects, segment drivable surfaces, or predict how other agents will move, while classical planners choose safe trajectories through that representation.
Public milestones helped prove the concept was more than laboratory demos. The DARPA Grand Challenge races in the mid-2000s pushed teams to cross desert terrain without human intervention; Stanford’s Stanley won the 2005 event after no vehicle finished the 2004 race. That result accelerated interest in perception stacks and onboard mapping for real-world robotics. See the DARPA Grand Challenge article and timeline entry.