Auto-Tunable Path Control for Dynamic Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path planning systems for autonomous vehicles struggle to efficiently navigate around obstacles and maintain a desired path while adapting to dynamic conditions.
Innovation Solution
A method and system for planning a path for autonomous vehicles that involves receiving vehicle position, velocity, and yaw, trimming a target path, creating graft paths, optimizing them to form a single graft path, and determining command steer angles and velocities to ensure the vehicle follows the path while avoiding obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the autonomous vehicle uses a fixed path planning system, then the system complexity is reduced, but the adaptability to dynamic obstacles and changing conditions deteriorates
Solution Approach 1:
The path planning system dynamically adjusts the target path based on real-time vehicle position, velocity, and yaw data. The system continuously regenerates graft paths as the vehicle moves, allowing the path to adapt dynamically to changing conditions and obstacles rather than following a static predetermined route
Solution Approach 2:
The system uses feedback from the vehicle's current state (position, velocity, yaw) to continuously adjust and optimize the target path. The path planning incorporates feedback loops that regenerate graft paths based on the vehicle's progress and environmental conditions, enabling adaptive response to dynamic obstacles
2Adaptability or versatility
If the system creates multiple graft paths for obstacle avoidance, then the adaptability to avoid obstacles improves, but the computational complexity and processing time increase
Solution Approach 1:
The path planning system segments the overall path into multiple graft paths that can be independently evaluated and optimized. Each graft path represents a potential route segment from the vehicle's current position to the target, allowing the system to create and compare multiple avoidance options without overwhelming computational complexity
Solution Approach 2:
The system generates multiple graft paths (excessive action) to ensure adequate obstacle coverage, but then optimizes them to select the most appropriate path. This approach ensures sufficient obstacle avoidance options are explored while the optimization process filters down to the best solution, balancing comprehensiveness with computational efficiency
3Manufacturing precision
If the vehicle follows a predetermined target path, then the path accuracy is maintained, but the ability to respond to dynamic obstacles deteriorates
Solution Approach 1:
The system maintains path accuracy by dynamically regenerating graft paths that are continuously optimized based on the vehicle's current state and environmental conditions. Rather than following a static predetermined path, the system adapts the target path in real-time while maintaining precision through continuous optimization of the graft path geometry
Solution Approach 2:
The system performs preliminary path planning by generating multiple graft paths before the vehicle reaches potential obstacle zones. This preliminary action allows the system to pre-compute avoidance routes while maintaining accuracy, so when obstacles are detected, the vehicle can quickly switch to pre-planned alternative paths without sacrificing precision
Data Source
AI summary
A path control systems and methods are disclosed that can aide in keeping an autonomous vehicle on a path. A current estimate of the autonomous vehicle's first position (x, y), heading, and velocity as well as a path of interest (e.g., breadcrumbs, line-arcs, or clothoid segments each containing velocity information) can be used to output a command velocity, command curvature, and/or a vector of waypoints. The command velocity, command curvature, and/or a vector of waypoints can be followed to move the autonomous vehicle from the first position onto the path of interest. An obstacle map may also be used to provide a path to the path of interest that avoids the obstacles on the obstacle map.