Occupancy Map Control for Dynamic Obstacle Noise in Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in accurately distinguishing between static and dynamic obstacles using point cloud data, leading to unnecessary path deviations due to point cloud noise from dynamic obstacles, which can result in inefficient navigation.
Innovation Solution
A moving object control system that identifies dynamic obstacles using captured images and depth information, generating an occupancy map with higher forgetting rates for dynamic obstacles, allowing for more accurate path planning and reducing the influence of point cloud noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If point cloud data is used to detect obstacles around a moving object, then obstacle detection capability is improved, but point cloud noise from dynamic obstacles causes unnecessary path deviations
Solution Approach 1:
The patent segments the obstacle detection and tracking process into distinct modules: point cloud data acquisition, dynamic obstacle identification using captured images, occupancy map generation with differentiated forgetting rates, and path planning. This segmentation allows each module to specialize in specific tasks, improving overall system reliability by isolating the effects of point cloud noise to specific processing stages.
Solution Approach 2:
The patent implements dynamic adjustment of forgetting rates in the occupancy map based on obstacle type. Dynamic obstacles (such as bicycles) have higher forgetting rates than static obstacles, allowing the system to adaptively respond to changing environmental conditions. This dynamic approach enables the system to reduce the influence of point cloud noise from dynamic obstacles while maintaining accurate detection of static obstacles.
2Device complexity
If a unified forgetting rate is used for all obstacles in the occupancy map, then system complexity is reduced, but dynamic obstacles cause unnecessary path deviations due to accumulated point cloud noise
Solution Approach 1:
The patent applies different forgetting rates to different types of obstacles within the occupancy map. Specifically, dynamic obstacles (identified through image recognition) receive a first forgetting rate, while static obstacles receive a second forgetting rate. This local differentiation allows the system to maintain simplicity in overall structure while achieving nuanced, context-appropriate behavior for different obstacle types, thereby improving navigation accuracy without excessive complexity.
3Area of stationary object
If point cloud noise from dynamic obstacles is accumulated in the occupancy map, then obstacle detection coverage is improved, but unnecessary travel path generation occurs
Solution Approach 1:
The patent implements dynamic forgetting rates that adjust based on obstacle characteristics. For dynamic obstacles like bicycles, a higher forgetting rate causes the occupancy map to naturally decay the noise accumulation over time, preventing permanent false obstacles from forming in the path planning data. This dynamic approach maintains adequate detection coverage while improving travel efficiency by reducing unnecessary path deviations.
Data Source
AI summary
A moving object control system acquires a captured image, captured by a moving object, and depth information of an environment captured in the captured image, identifies a dynamic obstacle that is autonomously movable and included in the captured image by using the captured image, and generates an occupancy map indicating occupancy of an obstacle for each divided region obtained by dividing a peripheral region of the moving object, on a basis of the depth information. The occupancy map includes a first divided region indicating occupancy of the dynamic obstacle forgotten according to a forgetting rate higher than a forgetting rate of a static obstacle that does not autonomously move. The system sets the divided region in a depth direction that is away from the moving object from a position of the identified dynamic obstacle, as the first divided region.