Graph-Based Path Planning Reducing Calculation Load
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Graph-based path planning for autonomous and semi-autonomous vehicles determines many more possible future locations than needed, leading to calculation-intensive processes and inefficiencies due to the need for frequent updates, while maintaining reliability and confidence.
Innovation Solution
A method that calculates projected host vehicle and obstacle locations using kinematic functions, determines collision potentials, and selectively generates qualified paths within a predetermined radius, ignoring duplicates and adjusting speed according to velocity profiles, to minimize the number of potential paths while ensuring safety and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If graph-based path planning calculates all possible future vehicle locations, then path planning completeness is improved, but calculation time and processing load increase significantly
Solution Approach 1:
The patent segments the continuous space of possible vehicle locations into discrete grid cells, and further segments the planning process into hierarchical levels (graph-based high-level planning followed by low-level trajectory generation). This segmentation allows the system to evaluate only relevant discrete locations rather than all continuous possibilities, reducing calculation time while maintaining planning completeness.
Solution Approach 2:
The patent applies partial action by calculating only the necessary subset of future locations required for safe path planning. Instead of computing all possible locations within the prediction horizon, the system uses the vehicle's current state (position, velocity, acceleration) and kinematic constraints to generate only the reachable and relevant future states, eliminating redundant calculations while ensuring sufficient coverage for safety.
2Reliability
If graph-based path planning evaluates many possible future locations, then path safety is improved, but computational complexity increases
Solution Approach 1:
The patent changes the parameters used to represent vehicle states from continuous values to discrete grid cell indices. By transforming the state representation and using discrete math instead of continuous optimization, the system reduces computational complexity while maintaining the ability to evaluate path safety through collision detection algorithms operating on the discrete grid.
3Measurement precision
If all possible future locations are calculated frequently, then real-time path accuracy is improved, but processing load increases
Solution Approach 1:
The patent performs preliminary action by pre-computing the graph structure and kinematic constraints before real-time operation. The graph representing possible transitions between grid cells is constructed offline or semi-offline, allowing real-time path planning to simply query and evaluate existing graph structures rather than computing all kinematic possibilities from scratch at each time step, thus reducing processing load while maintaining accuracy.
Data Source
AI summary
A method of path planning for a host vehicle includes: receiving host vehicle, environmental and obstacle information; calculating one or more projected host vehicle locations; computing a projected obstacle location for each obstacle; and determining a collision potential between each projected host vehicle location and each projected obstacle location. Until a maximum number of steps is reached, and while at least one projected host vehicle location has an associated collision potential below a collision threshold, the method further includes repeating the calculating, computing and determining steps.