Occupancy Map Control With Region-Based Obstacle Forgetting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing micro-mobility vehicle systems face challenges in efficiently managing obstacle detection and path planning, particularly in setting a suitable forgetting rate for obstacle information across different regions around the moving object.
Innovation Solution
A moving object control system that accumulates obstacle information for each divided region, sets a predetermined forgetting rate for each region, and generates an occupancy map to guide path planning, allowing for dynamic adjustment of forgetting rates based on the region's proximity to the viewing angle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If obstacle information is accumulated for all regions around the moving object, then path planning accuracy is improved, but memory usage and processing complexity increase
Solution Approach 1:
The peripheral region is divided into multiple regions based on distance from the moving object. Different forgetting rates are applied to each region, allowing the system to maintain high path planning accuracy by remembering obstacles in critical near regions while reducing processing complexity by applying faster forgetting to distant regions.
Solution Approach 2:
Different forgetting rates are assigned to different regions based on their importance. Regions closer to the moving object (more critical for path planning) have lower forgetting rates to retain obstacle information longer, while distant regions have higher forgetting rates. This local differentiation optimizes both accuracy and complexity.
2Reliability
If obstacle information is retained for longer periods, then path avoidance capability is improved, but the system becomes less responsive to new obstacles
Solution Approach 1:
The forgetting rate is made dynamic and region-dependent rather than uniform. By applying different forgetting rates to different regions, the system maintains reliability for path avoidance in critical areas while remaining responsive to new obstacles in less critical areas, achieving a balance between retention and adaptability.
Solution Approach 2:
The system pre-establishes region-based forgetting rate policies before obstacle detection. This allows the system to automatically apply appropriate retention or forgetting behavior based on the obstacle's location, ensuring both reliable avoidance capability and responsiveness without real-time complex decision-making.
3Device complexity
If a single sensor is used to detect the front periphery, then hardware resources are reduced, but detection coverage is limited
Solution Approach 1:
The system compensates for the limited spatial coverage of a single sensor by extending detection in the time dimension. By accumulating obstacle information over time with region-based forgetting rates, the system effectively expands its operational awareness beyond the immediate sensor field of view, achieving comprehensive coverage through temporal integration.
Data Source
AI summary
A moving object control system of the present invention comprises accumulating information on an obstacle detected in a past for each divided region obtained by dividing peripheral regions of a moving object; causing information on the accumulated obstacle to be forgotten in accordance with a predetermined forgetting rate allocated to each divided region; acquiring a captured image; detecting an obstacle included in the captured image; and generating an occupancy map indicating occupancy of an obstacle for each divided region in accordance with information on the accumulated obstacle and the detected obstacle for a current peripheral region of the moving object.