Vehicle Trajectory Tree Expansion for Low-Latency Path Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing trajectory planning methods for autonomous vehicles are computationally intensive and inefficient, leading to high latency and resource usage, particularly in dynamic environments where exhaustive tree search is required to predict future object trajectories accurately.
Innovation Solution
A selective expansion technique for tree structures is employed, focusing on higher likelihood scenarios by prioritizing the exploration of more informative branches and pruning redundant ones, using similarity and quality metrics to determine which state sibling sets to expand.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If exhaustive tree search is used to predict future object trajectories, then prediction accuracy is improved, but computational cost and processing time increase significantly
Solution Approach 1:
The patent segments the tree search process by identifying and separating high-probability trajectory branches from low-probability ones. The system divides the computation into essential expansions (high likelihood scenarios) and optional expansions (low likelihood scenarios), allowing the planner to focus computational resources on the most relevant predictions while skipping unnecessary calculations.
Solution Approach 2:
The patent applies partial action by performing only the necessary tree expansions required to achieve sufficient prediction accuracy. Instead of exhaustively expanding all possible trajectories, the system expands only those branches with probability above a threshold, accepting that some low-probability scenarios may remain unexplored, thus reducing computation time while maintaining practical accuracy.
2Measurement precision
If exhaustive tree search is used to predict future object trajectories, then prediction accuracy is improved, but computational resource usage increases
Solution Approach 1:
The patent segments the computation into high-priority and low-priority trajectory evaluations. By identifying that only high-probability branches contribute significantly to accurate prediction, the system separates these from low-probability branches that consume resources but provide minimal accuracy improvement, enabling selective computation that optimizes the accuracy-resource tradeoff.
Solution Approach 2:
The patent changes the probability threshold parameter dynamically based on computational constraints and environmental factors. By adjusting this parameter, the system can control the balance between prediction accuracy and resource consumption, expanding more branches when resources are abundant and fewer branches when resources are constrained, thus adapting computational effort to available resources.
3Reliability
If all state sibling sets are expanded in the tree structure, then trajectory planning completeness is improved, but memory usage and computational overhead increase
Solution Approach 1:
The patent segments the state sibling sets into groups based on their probability characteristics and relevance to the current planning context. Instead of uniformly expanding all sibling sets, the system identifies and expands only those groups that are most likely to lead to valid trajectories, pruning groups with very low probabilities, thus reducing tree complexity while preserving planning completeness for relevant scenarios.
Solution Approach 2:
The patent performs preliminary evaluation of state sibling sets before full expansion, using probability estimates and heuristic assessments to identify which sets warrant expansion. This preliminary filtering action occurs before the main tree expansion process, allowing the system to avoid committing computational resources to unlikely scenarios while ensuring that promising branches are fully explored.
Data Source
AI summary
Techniques for determining a vehicle trajectory that causes a vehicle to navigate in an environment relative to one or more objects are described herein. In some cases, the techniques described herein relate to selectively expanding a tree structure (e.g., a decision tree structure) to efficiently search for simulation data that can be used to evaluate vehicle control trajectories. The tree structure may include state nodes representing observed and/or predicted environment states, and action nodes representing candidate actions the vehicle may take. By selectively and incrementally expanding the tree using estimated state transition probabilities to focus on higher likelihood scenarios, more optimal trajectories can be determined without exhaustively evaluating every possible outcome.