Spatial Graph Pruning for Human-Like Vehicle Path Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicle systems face challenges in generating high-quality navigation paths due to the computational complexity associated with dense graph representations of spatial structures, which can lead to inefficient path planning and increased processing time.
Innovation Solution
A driving data-guided spatial planning system that reduces the complexity of spatial graphs by labeling and pruning edges based on distance metrics and human driving data, focusing on useful edges to replicate human-level performance while optimizing computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If dense graph representations of spatial structures are used, then path planning accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent extracts and removes unnecessary edges from the spatial graph based on human driving data analysis. By identifying and eliminating edges that do not correspond to actual human driving patterns, the system reduces graph density while preserving path planning accuracy for realistic scenarios.
Solution Approach 2:
The patent applies different levels of graph density to different spatial regions. Areas with high human driving activity retain detailed graph representations, while less critical areas use coarser representations, optimizing the balance between accuracy and computational complexity locally.
2Manufacturing precision
If dense graph representations of spatial structures are used, then path planning quality is improved, but processing time increases
Solution Approach 1:
The patent removes redundant edges from the graph that do not contribute to human-like path planning. This extraction reduces the number of computations required during path planning while maintaining the quality of paths that replicate human driving behavior.
Solution Approach 2:
The patent performs preliminary analysis of human driving data before path planning to identify and pre-remove unnecessary edges. This preprocessing step reduces the computational burden during actual path planning operations, decreasing processing time while preserving path quality.
3Adaptability or versatility
If all edges in spatial graph are retained, then coverage of possible paths is improved, but computational efficiency decreases
Solution Approach 1:
The patent applies selective edge retention based on local driving patterns. Edges that correspond to common human driving behaviors are retained, while rare or unrealistic paths are removed, maintaining adaptability for realistic scenarios while improving computational efficiency.
Solution Approach 2:
The patent extracts and removes edges that do not represent plausible human driving paths. By eliminating these unnecessary edges while preserving those that match human behavior patterns, the system maintains adequate coverage of possible paths while significantly improving computational efficiency.
Data Source
AI summary
Among other things, techniques are described for driver data guided spatial planning. A spatial structure is generated comprising a plurality of nodes connected by edges. At least some of the nodes and edges represent a path to navigate a vehicle from a first point to a second point. Edges of the spatial structure are labeled as useful based on a distance metric. The spatial structure is pruned by removing one or more edges from the spatial structure according to a respective label of the edges, wherein an extent of the removal is based on a predetermined graph size, a predetermined performance, or any combinations thereof to obtain a pruned graph. A path from the first point to the second point on the pruned graph is identified and the vehicle is navigated in accordance with the path from the first point to the second point on the pruned graph.