Autonomous Vehicle Trajectory Planning via Station-Time Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving vehicle (ADV) trajectory generation systems face challenges in safely and efficiently navigating through environments with obstacles while adhering to traffic rules, as they struggle to balance safety, comfort, and compliance with traffic regulations.
Innovation Solution
The system projects obstacles onto a station-time graph, determines end points that avoid these obstacles, generates trajectory candidates based on these end points, and selects a trajectory using a predetermined algorithm to ensure safe and efficient navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the trajectory generation process prioritizes safety by avoiding obstacles, then collision risk is reduced, but the navigation time and efficiency may increase due to longer paths
Solution Approach 1:
The patent transforms the traditional spatial trajectory planning into a station-time graph dimensionality, where trajectories are planned in the (station, time) domain rather than just spatial domain. This allows the system to visualize and optimize both safety (avoiding obstacle regions in the graph) and efficiency (minimizing time dimension) simultaneously by selecting optimal end points and generating trajectory candidates that balance both objectives.
2Ease of operation
If the trajectory generation process ensures smooth and comfortable motion, then passenger comfort is improved, but the time to reach destination may increase due to graceful accelerations
Solution Approach 1:
The patent generates multiple trajectory candidates with different motion characteristics (acceleration profiles, speed variations) and selects the optimal trajectory based on comfort and efficiency criteria. By adjusting motion parameters such as acceleration and velocity within the trajectory generation process, the system can balance comfort (graceful accelerations) and travel time (efficient navigation) according to different operational requirements.
3Reliability
If the trajectory generation process strictly follows traffic rules, then compliance is improved, but the flexibility in route selection is reduced
Solution Approach 1:
The patent incorporates dynamic traffic rule constraints into the trajectory generation process, where traffic signals and rules are integrated as time-dependent constraints in the station-time graph. The system can adaptively adjust trajectory candidates based on real-time traffic conditions while maintaining compliance, allowing flexibility in route selection within the bounds of traffic regulations rather than rigidly following predetermined paths.
Data Source
AI summary
In one embodiment, in response to detecting an obstacle based on a driving environment surrounding an autonomous driving vehicle (ADV), a system projects the obstacle onto a station-time (ST) graph, where the ST graph indicates a location of the obstacle relative to a current location of the ADV at different points in time. The system determines a first set of end points that are not overlapped with the obstacle within the ST graph, wherein each of the end points in the first set represents a possible end condition. The system generates a first set of trajectory candidates between a starting point representing the current location of the ADV and the end points of the first set based on the ST graph. The system selects one of the trajectory candidates in the first set using a predetermined trajectory selection algorithm to control the ADV in view of the obstacle.