Path Curve Cost Design for Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for calculating the cost of path curves for autonomous driving vehicles are inadequate in representing complex environments and are computationally expensive, leading to undesired outputs and limitations in path updates.
Innovation Solution
A method is introduced to determine a least-cost path curve by generating candidate path curves with attributes such as obstacle, physical lane, virtual lane, and safety-comfort categories, using logical operations and priority-based comparisons to select the optimal path for autonomous navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional single-value cost calculation is used, then computation is simpler, but the cost cannot represent complex environments accurately
Solution Approach 1:
The cost calculation is segmented into multiple discrete components: obstacle cost, lane boundary cost, and safety-comfort cost. Each component is calculated separately based on specific criteria (obstacle proximity, lane boundary proximity, curvature, jerk) and then aggregated. This segmentation allows the system to accurately represent complex environments while maintaining computational efficiency through modular calculation.
2Productivity
If traditional cost calculation methods are used, then the system is easier to implement, but computational cost is too high for real-time updates
Solution Approach 1:
By dividing the cost calculation into separate modular components (obstacle cost, lane boundary cost, safety-comfort cost), the system can efficiently compute each component independently and aggregate them. This modular approach reduces overall computational complexity and enables real-time path updates at higher frequencies.
Solution Approach 2:
The patent transforms the cost representation from a single continuous value to a structured set of discrete parameters with associated weights. This parameterization allows for more efficient computation through standardized calculation formulas and enables dynamic adjustment of priority weights without requiring complete recalculation of the cost model.
3Manufacturing precision
If more path points are used to determine cost, then path accuracy improves, but computational expense increases
Solution Approach 1:
The patent calculates costs at discrete path points along the candidate curves rather than continuously, segmenting the computation into manageable sampling points. This approach maintains path planning precision by evaluating costs at sufficient intervals while significantly reducing computational energy requirements compared to continuous evaluation.
Data Source
AI summary
A new cost design is disclosed for evaluating candidate path curves for navigating an autonomous driving vehicle (ADV) through a segment of a route which may include an obstacle. Each point on each candidate path curve has a plurality of attributes having logical values and an associated priority of evaluation, and at least one numeric attribute having an associated priority of evaluation. A cost for each path curve is determined using the attributes and priorities, and a least cost path curve is selected using the attributes and priorities. By comparing attribute values in accordance with priority, and utilizing logical values, the efficiency of determining path curve cost and selecting a least cost path curve is substantially improved.