Autonomous Vehicle Trajectory Planning for Wait-or-Replan Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motion planning methods for autonomous driving vehicles face challenges in complex scenarios such as parking, three-points-turns, and obstacle avoidance, especially in road driving scenarios requiring lane boundaries, and free space driving scenarios without lane boundaries, where they are slow to generate real-time trajectories and perform poorly in obstacle avoidance.
Innovation Solution
A method that determines the probability of obstacle residence time using a probability density function, allowing the vehicle to decide whether to wait or re-plan a trajectory based on estimated arrival times, and uses modified A-star searching algorithms for real-time trajectory planning in both on-lane and open-space modes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional motion planning methods are used for road driving scenarios with lane boundaries, then the vehicle can navigate following defined paths, but the method is difficult to deal with complex scenarios such as parking, three-points-turn and obstacles avoidance
Solution Approach 1:
The patent segments the motion planning problem into two distinct modes: road driving mode (with lane boundaries) and free space driving mode (without lane boundaries). Each mode has its own specialized planning algorithm, allowing the system to handle complex scenarios by switching between modes rather than using a single complex planner for all situations.
Solution Approach 2:
The system dynamically switches between different motion planning methods based on the current driving scenario. When lane boundaries are detected, the road driving planner is activated; when in free space scenarios (parking, three-points-turn), the system transitions to the free space planner, making the system adaptable without requiring a permanently complex structure.
2Adaptability or versatility
If conventional free space path planning method is used, then the vehicle can navigate in open spaces without lane boundaries, but the method is slow to generate a trajectory in real-time
Solution Approach 1:
The patent separates free space path planning into a dedicated module that activates only when lane boundaries are absent. This segmentation allows the system to use specialized algorithms (like probabilistic road maps or RRT*) for free space scenarios without impacting the real-time performance requirements of road-driven scenarios, where more efficient lane-following algorithms are used instead.
3Adaptability or versatility
If conventional free space path planning method is used, then the vehicle can operate without lane boundaries, but the method may lead to a poor performance in obstacle avoidance
Solution Approach 1:
The system continuously monitors the driving environment through sensor input to detect obstacles and determine whether the vehicle is in a road driving scenario or free space scenario. This feedback loop allows the system to switch between planning modes and adjust trajectories in real-time to avoid obstacles, improving reliability by actively responding to environmental conditions rather than relying on pre-planned static paths.
Data Source
AI summary
A computer-implement method for operating an ADV is disclosed. A first trajectory for the ADV to drive along is planned. The ADV is to autonomously drive along the first trajectory. An obstacle in an affected region of the ADV is detected based on sensor data obtained from a plurality of sensors mounted on the ADV. An expected residence time of the obstacle in the affected region is determined. Whether to plan a second trajectory or to wait for the obstacle to leave the affected region is determined based on the expected residence time of the obstacle in the affected region. A second trajectory for the ADV to drive along is planned and the ADV is to autonomously drive along the second trajectory, or the ADV is to wait for the obstacle to leave the affected region and to autonomously drive along the first trajectory afterwards.