A decision-oriented trajectory planning approach with longitudinal priority
2022-01-0095
03/29/2022
- Event
- Content
- Most optimizations for trajectory planning with autonomous vehicles first decouple the lateral-longitudinal-time problem, transforming decision-making into a matter of limited choices. Potential alternatives that can lead to better performance have been ruled out once the trajectory has been predefined in one or two dimensions, such as path optimization. This article mainly draws on the optimization method from the Apollo EM Motion Planner and proposes an optimal approach in Frenét-Frame by putting the decision-making process as well as the speed optimization first. The approach mainly concentrates on multilane scenarios with high-speed autonomous driving in a two-layer manner. The first layer consists of decision-maker and converter, giving decision-based obstacle boundaries as well as short-term objectives to the trajectory generator. The generator in the second layer puts the longitudinal axis first, computing in parallel with every decision to acquire the best trajectory, whose average cost has the minimum. Compared to the previous method, less dynamic programming and quadratic programming have been applied for optimization since shorter unit sampling time and dynamic constraints are implemented. Decision-oriented planning makes up the possible decision loss at the beginning and suits well with motorway and multilane scenarios. Since the short-term decision is settled, the planning order with longitudinal priority simplifies the dynamic confirmation of obstacles and reduces choices in the non-convex map. The decision-converter is a must to avoid unsuccessful trajectories. Trajectory generation safety depends primarily on the appropriate number of sampling points in the longitudinal axis while the obstacle prediction determines driving effectiveness. The capabilities of the approach are testified under hundreds of virtual high-speed motorway scenarios with some random character.
- Citation
- Guo, R., and Yu, S., "A decision-oriented trajectory planning approach with longitudinal priority," SAE Technical Paper 2022-01-0095, 2022, .