Autonomous Vehicle Path Planning With Dynamic Step-Size Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path planning algorithms for autonomous vehicles face challenges in providing real-time dynamic solutions, consuming excessive computing resources, and failing to optimize paths for collision-free, kinematic feasibility, and minimal length.
Innovation Solution
An environment-aware path planning system using a dynamic step-size generator to generate a sequence of nodes for collision-free and kinematically-feasible paths, utilizing sensors and a computing device to create an occupancy grid map and select the least path length among candidate sub-paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional path planning algorithms are used, then path planning can be performed, but computation speed is slow and large amount of computing resources are consumed
Solution Approach 1:
The path planning algorithm is segmented into multiple layers: global path planning determines the overall route, while local path planning handles real-time obstacle avoidance. This segmentation allows each layer to operate with appropriate computational complexity, reducing overall computing resource consumption while maintaining planning speed.
Solution Approach 2:
The algorithm dynamically adjusts the search radius and planning horizon based on vehicle speed, obstacle density, and environmental complexity. When obstacles are sparse or vehicle speed is high, the search radius is reduced, significantly decreasing computational load while maintaining safety and optimality.
2Reliability
If real-time obstacle avoidance is implemented, then collision-free paths can be achieved, but dynamic solution and path updating cannot be carried out simultaneously in real-time
Solution Approach 1:
The global path planner pre-computes the overall route and key waypoints before the vehicle encounters obstacles. This preliminary path provides a foundation that reduces the computational burden during real-time local planning, enabling faster path updates while maintaining collision-free navigation.
Solution Approach 2:
The path planning system operates continuously with overlapping global and local planning processes. While the global planner updates the overall route, the local planner simultaneously handles real-time obstacle avoidance. This continuous operation ensures seamless path updates without interruption, maintaining both reliability and real-time performance.
3Reliability
If optimal paths are planned considering multiple constraints, then collision-free and kinematically-feasible paths are achieved, but computation speed decreases
Solution Approach 1:
The algorithm applies different levels of constraint checking and optimization to different regions of the search space. Areas with high obstacle density or complex kinematic requirements receive more rigorous evaluation, while open areas use simplified assessment. This local quality approach maintains kinematic feasibility where needed while improving overall planning efficiency.
Data Source
AI summary
A system and method of planning a path for an autonomous vehicle from the vehicle's initial configuration to the goal configuration which is collision-free, kinematically-feasible, and near-minimal in length. The vehicle is equipped with a plurality of perception sensors such as lidar, camera, etc., and is configured to operate in an autonomous mode. An onboard computing device is configured to process the sensor data and provide a dynamic occupancy grid map of the surrounding environment in real-time. Based on the occupancy grid map, the path planner can quickly calculate a collision-free and dynamically feasible path towards the goal configuration for the vehicle to follow.