Autonomous Vehicle Trajectory Planning With Hierarchical Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous navigation systems face challenges in quickly planning and re-planning vehicle trajectories to optimize travel time and safety, due to the need to account for new sensor data, obstacles, and dynamic vehicle conditions.
Innovation Solution
A tripartite path-planning system is employed, consisting of a global planner, a fast local planner with a simplified vehicle dynamics model, and a local planner with an advanced vehicle dynamics model. This system rapidly calculates and updates optimal trajectories, ensuring safety and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If trajectory is recalculated each time new sensor data is received to ensure reliability, then safety is improved, but computational overhead increases and processing speed decreases
Solution Approach 1:
The path planning system is divided into three hierarchical levels: global planner (long-term route), fast local planner (medium-term trajectory with simplified dynamics), and local planner (short-term control with advanced dynamics). This segmentation allows different planning frequencies and computational complexities at each level, enabling reliable trajectory recalculation without overwhelming computational resources.
Solution Approach 2:
The system dynamically adjusts the vehicle dynamics model complexity based on planning needs. The fast local planner uses a simplified dynamics model for quicker calculations, while the local planner uses an advanced dynamics model for precise control. This dynamic adaptation of model complexity optimizes the balance between calculation speed and accuracy.
2Manufacturing precision
If advanced vehicle dynamics model is used in local planner to ensure safety within handling limits, then trajectory accuracy and safety are improved, but computational complexity increases
Solution Approach 1:
The planning system segments computational tasks across three levels. The local planner with advanced dynamics model focuses only on short-term control (last level of planning), while global and fast local planners handle higher-level route planning. This segmentation limits the computational burden on any single planner while maintaining overall trajectory precision.
Solution Approach 2:
The advanced vehicle dynamics model is applied locally at the local planner level where it is most needed for precise control and safety assurance. The simplified model is used at the fast local planner level for quicker trajectory generation. This localized application of complexity ensures high precision where required without unnecessary computational overhead elsewhere.
Data Source
AI summary
A multi-layer path-planning system and method calculates trajectories for autonomous vehicles using a global planner, a fast local planner, and an optimizing local planner. The calculated trajectories are used to guide the autonomous vehicle along a bounded path between a starting point and a destination.