Autonomous Vehicle Trajectory Planning Under Localization Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face safety issues due to localization uncertainty caused by sensor inaccuracy and speed uncertainty, which negatively impact trajectory planning and vehicle control.
Innovation Solution
A method that determines a confidence region based on localization and speed uncertainty, allowing for probability-based trajectory planning and control to avoid collisions, by calculating a circular area around the vehicle's potential location and adjusting control commands to minimize collision risk.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional localization methods are used without considering uncertainty, then the system is simpler to implement, but safety and collision avoidance capability deteriorate
Solution Approach 1:
The system performs preliminary calculation of confidence regions based on localization uncertainty before trajectory planning and control decisions are made. By pre-characterizing the uncertainty envelope around the vehicle position, the system enables safety-aware planning without adding complexity to the core localization module, thus improving reliability while maintaining manageable system complexity.
Solution Approach 2:
The confidence region acts as an intermediary between the localization module and the trajectory planning module. Instead of directly handling complex uncertainty propagation through the entire control stack, the system introduces this intermediate representation that encapsulates localization uncertainty in a geometric form, making it easier for higher-level modules to make safety decisions.
2Reliability
If confidence region calculation is performed, then collision avoidance capability is improved, but computational load increases
Solution Approach 1:
The system changes the parameter representation of localization uncertainty from complex probability distributions to simplified geometric confidence regions (ellipses or circles). This parameter transformation maintains the essential uncertainty information needed for collision avoidance while significantly reducing the computational complexity of subsequent trajectory planning and safety assessments.
Solution Approach 2:
The system calculates confidence regions with a level of detail sufficient for collision avoidance purposes, rather than performing complete and exact uncertainty propagation. By computing only the essential geometric boundaries needed for safety decisions, the system achieves adequate collision avoidance capability without the full computational burden of rigorous probabilistic methods.
3Measurement precision
If localization uncertainty is incorporated into trajectory planning, then navigation accuracy is improved, but planning complexity increases
Solution Approach 1:
The confidence region representing localization uncertainty is calculated in advance and made available to the trajectory planning module before path optimization begins. This preliminary preparation allows the planner to incorporate uncertainty constraints without performing complex real-time uncertainty propagation during the planning optimization process, thus improving navigation accuracy while limiting planning complexity.
Data Source
AI summary
Systems, methods, and media for factoring localization uncertainty of an ADV into its planning and control process to increase the safety of the ADV. The uncertainty of the localization can be caused by sensor inaccuracy, map matching algorithm inaccuracy, and/or speed uncertainty. The localization uncertainty can have negative impact on trajectory planning and vehicle control. Embodiments described herein are intended to increase the safety of the ADV by considering localization uncertainty in trajectory planning and vehicle control. An exemplary method includes determining a confidence region for an ADV that is automatically driving on a road segment based on localization uncertainty and speed uncertainty; determining that an object is within the confidence region, and a probability of collision with the ADV based on a distance of the object to the ADV; and planning a trajectory based on the probability of collision, and controlling the ADV based on the probability of collision.