Automated driving technology has garnered increasing attention due to its potential to reduce traffic accidents, alleviate congestion, and enhance travel convenience. As a key component of automated driving technology, trajectory planning plays a pivotal role in determining feasible trajectories in different traffic scenarios. However, there is currently insufficient research on the performance of different curve models in lane-changing scenarios. Furthermore, the conventional speed planning method (DP&QP) often ignores vehicle kinematics and traffic scenarios when generating convex spaces, resulting in inefficiency and longer run time. In path planning, this paper compared the performance of Dubins, Sine, Bézier, and B-spline curves in lane change scenarios, ultimately selecting the B-spline curve as the lane change path model to ensure that the vehicle remains close to the center line of the road after a lane change. In speed planning, we introduce a novel method for directly generating a convex space, taking into account the kinematics of the ego vehicle and the current traffic conditions. This approach allows for adaptive adjustments to the speed planning scheme based on the traffic situation. Compared with the conventional trajectory planning method, this approach significantly reduces the run time by 32.7%.
Faster Trajectory Planning for Lane Change Scenarios with Dynamic Environment
12.01.2024
699043 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
A Dynamic Lane-Change Trajectory Planning Algorithm Based on Minimum Safe Spacing
SAE Technical Papers | 2021
|A Dynamic Lane-Change Trajectory Planning Algorithm Based on Minimum Safe Spacing
British Library Conference Proceedings | 2020
|A Dynamic Lane-Change Trajectory Planning Algorithm Based on Minimum Safe Spacing
British Library Conference Proceedings | 2020
|