Motion-Primitive Path Planning on Occupancy Grids for Unstructured Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face navigation challenges when pathway indicators such as runway lights, lane markings, and traffic signs are obscured, degraded, or invalidated, leading to inaccurate and inefficient navigation in unstructured environments.
Innovation Solution
An autonomous vehicle guidance system that generates paths based on an occupancy map, using sensor data to detect occupied and free spaces, and precomputed motion primitives and cost plots to determine collision-free and low-cost paths, even in the absence of traditional indicators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous vehicles rely on traditional pathway indicators (runway lights, lane markings, traffic signs), then navigation is simple and direct in structured environments, but navigation accuracy and reliability deteriorate when indicators are obscured, degraded, or invalidated
Solution Approach 1:
The system uses sensor data (lidar, cameras, radar) to perform multiple functions: detecting occupied spaces, generating occupancy maps, and identifying pathways simultaneously. This multi-functional approach allows the vehicle to navigate both structured environments with traditional indicators and unstructured environments without clear markings, resolving the contradiction between reliability in structured settings and adaptability to unstructured settings.
Solution Approach 2:
The patent introduces an occupancy map as an intermediary representation between raw sensor data and navigation decisions. The occupancy map abstracts the environment into occupied and free spaces, allowing the vehicle to navigate based on this intermediate representation rather than directly relying on traditional pathway indicators. This mediator enables reliable navigation in both structured and unstructured environments.
2Measurement precision
If the system generates paths continuously using sensor data and occupancy maps, then navigation accuracy improves in unstructured environments, but computational complexity and processing time increase
Solution Approach 1:
The system precomputes motion primitives (pre-defined motion sequences) and cost plots (cost-to-go maps) before real-time path generation. These precomputed elements are stored and reused during actual navigation, reducing the computational burden during critical path generation. This allows the system to maintain high path generation accuracy while managing computational complexity through advance preparation.
Solution Approach 2:
The path generation process is segmented into distinct components: sensor data acquisition, occupancy map generation, motion primitive selection, and path assembly. By dividing the complex path generation task into these manageable segments, the system can process each component separately using optimized algorithms, improving overall accuracy while controlling computational complexity through modular processing.
3Speed
If the vehicle uses precomputed motion primitives and cost plots for path generation, then path computation speed improves, but adaptability to dynamic obstacles and changing environments may be reduced
Solution Approach 1:
The system dynamically selects and adjusts motion primitives based on real-time occupancy map updates and current vehicle state. Rather than using fixed precomputed paths, the system chooses from multiple motion primitives the ones most appropriate for current conditions, and continuously updates the occupancy map to reflect dynamic obstacles. This dynamic approach maintains fast computation speeds while adapting to changing environments.
Solution Approach 2:
The system implements feedback loops where the occupancy map is continuously updated with new sensor data, and path generation uses this updated information to select appropriate motion primitives. The cost plots provide feedback on the optimality of different paths, allowing the system to adjust its trajectory in real-time based on dynamic obstacles while maintaining computational efficiency through the structured feedback mechanism.
Data Source
AI summary
An autonomous vehicle guidance system that generates a path for controlling an autonomous vehicle based at least in part on a data structure generated based at least in part on sensor data that may indicate occupied space in an environment surrounding an autonomous vehicle. The guidance system may receive a grid and generate a grid associated with the grid and the data structure. The guidance system may additionally or alternatively sub-sample the grid (latterly and/or longitudinally) dynamically based at least in part on characteristics determined from the data structure. The guidance system may identify a path based at least in part on a set of precomputed motion primitives, costs associated therewith, and/or a heuristic cost plot that indicates a cheapest cost to move from one pose to another.