Polyline Path Optimization for Autonomous Vehicle Obstacle Evasion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous and semi-autonomous vehicles face challenges in generating an optimal path that avoids obstacles while maintaining comfort and avoiding sudden turns or collisions, as existing methods often result in suboptimal paths that may cause occupant discomfort and increased risk of collisions.
Innovation Solution
A system and method that uses sensors to detect obstacles and lane boundaries, calculates lateral constraints, and employs a quadratic program to optimize the vehicle's path by adjusting y-coordinates of polyline nodes, ensuring the path avoids obstacles while minimizing sharp turns and maintaining proximity to the original path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the vehicle generates a path that follows the centerline position, then the path generation is simple, but the path may not avoid obstacles effectively
Solution Approach 1:
The path is represented as a polyline with multiple discrete nodes instead of a continuous curve. Each node can be independently optimized for obstacle avoidance while maintaining overall path continuity. This segmentation allows the system to balance simplicity in path representation with effectiveness in obstacle avoidance by adjusting individual node positions.
Solution Approach 2:
The path is continuously updated as the vehicle travels through the road segment. New obstacles are detected and incorporated into path recalculation, and the polyline nodes are dynamically adjusted to avoid obstacles while maintaining proximity to the original centerline path. This dynamic adaptation ensures both simplicity and effectiveness.
2Loss of time
If the vehicle quickly generates a new path to avoid obstacles, then the response time is reduced, but the path quality may deteriorate causing occupant discomfort
Solution Approach 1:
The optimization focuses specifically on adjusting the y-coordinates of polyline nodes rather than recalculating the entire path geometry. This parameter-specific optimization reduces computational complexity and generation time while still achieving effective obstacle avoidance. The quadratic program efficiently determines optimal y-coordinates that balance obstacle avoidance with smooth path transitions for occupant comfort.
Solution Approach 2:
The path optimization is applied locally to specific segments of the polyline rather than globally to the entire path. Only nodes near detected obstacles are adjusted, while other portions of the path remain unchanged. This local optimization reduces computational time and ensures that path modifications are minimal and smooth, maintaining occupant comfort while achieving timely obstacle avoidance.
3Ease of operation
If the vehicle maximizes comfort by avoiding sharp turns, then occupant comfort is improved, but the ability to quickly avoid obstacles is reduced
Solution Approach 1:
The system applies partial optimization by adjusting only the necessary portion of the path (y-coordinates of nodes near obstacles) rather than redesigning the entire path. This partial action achieves sufficient obstacle avoidance without introducing excessive path modifications that would cause sharp turns and occupant discomfort. The quadratic program determines the minimum necessary adjustments to achieve effective avoidance.
4Ease of operation
If the vehicle stays close to the original path, then path deviation is minimized, but obstacle avoidance effectiveness is reduced
Solution Approach 1:
The optimization operates in the lateral dimension (y-coordinates) while keeping the longitudinal dimension (x-coordinates) fixed along the original path. This dimensional approach allows the vehicle to stay close to the original path in terms of longitudinal position while achieving effective obstacle avoidance through lateral adjustments of polyline nodes. The quadratic program finds optimal y-coordinates that balance proximity to the original path with obstacle avoidance requirements.
Data Source
AI summary
In one embodiment, example systems and methods relate to generating an optimal path for a vehicle to avoid an obstacle. As the vehicle travels on a road segment, a polyline representing a path of the vehicle is generated. The polyline may include a plurality of nodes, and each node may have a set of coordinates with respect to the polyline. The vehicle uses sensors to detect one or more obstacles on the road, and lane boundaries of the road. The vehicle uses these obstacles and lane boundaries to calculate lateral constraints for the polyline. The vehicle then uses the lateral constraints and a quadratic program to optimize the polyline. The optimized polyline may be used to determine the optimal path for the vehicle that avoids the one or more obstacles.