Planning with dynamic state a trajectory of an autonomous vehicle

• Main Authors: Hans Andersen, Scott Drew Pendleton
• Format: Patent
• Language: English
• Published: 26.10.2022
• Subjects: CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE ORDIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; PERFORMING OPERATIONS; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TOTHE CONTROL OF A PARTICULAR SUB-UNIT; TRANSPORTING; VEHICLES IN GENERAL

Method of defining a trajectory for an autonomous vehicle (AV) 100 navigating around obstacles along a route 1402, by detecting objects proximate the route; identifying static constrains including non-temporal, time-independent risks, e.g. road work 1406, speed limit, or other stationary objects; estimating a preliminary trajectory; identifying dynamic constrains including temporal, time-dependent risks, e.g. traffic lights at a junction 1404, pedestrians 1504, surrounding vehicles, and adjusting and refining the trajectory to avoid both static and dynamic risk constrains. Predicted behaviour of objects and confidence in the prediction may influence the size of temporal zone 1506. The trajectory may be adjusted if the temporal zone is a threshold distance away. A risk may be prioritised over a second risk. Static constrains may be grouped into soft and hard constrains, and a dynamic constraint may allow a soft constrain to be violated to avoid temporal risks. Duration of amber traffic light, maximum acceleration and deceleration rate, and trailing vehicles may be taken into account when determining constrains.