Obstacle Placement Verification for Autonomous Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in achieving high accuracy due to discrepancies between symbolized and actual driving scenes, leading to potential errors in obstacle detection and navigation.
Innovation Solution
A vehicle periphery information verification device that includes a periphery information detection instrument, a host vehicle information detection instrument, and a controller, which arranges obstacles on a map to determine a travelable region and judges contradictions between detected obstacles and the region, preventing autonomous driving actions based on potentially inaccurate information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If obstacle information is arranged on a map for autonomous driving navigation, then the drive scene accuracy is improved, but detection errors may occur when obstacles are incorrectly positioned outside the travelable region
Solution Approach 1:
The system performs preliminary verification by checking whether detected obstacles are positioned within the travelable region before using them for autonomous driving decisions. The contradiction judgment unit checks obstacle positions against the map-based travelable region in advance, preventing erroneous obstacles from affecting navigation decisions.
Solution Approach 2:
The system implements a feedback mechanism where the contradiction judgment unit continuously monitors obstacle positions and provides feedback to the autonomous driving determination unit. When contradictions are detected (obstacles outside travelable region), the system adjusts its decision-making process by prohibiting actions based on erroneous information.
2Reliability
If the system prohibits autonomous driving actions when contradictions are detected, then safety is improved, but driving operations may be restricted unnecessarily
Solution Approach 1:
The system applies local quality by selectively prohibiting only those autonomous driving actions that would be based on contradictory obstacle information, while allowing other driving operations to proceed normally. The prohibiting unit targets specific problematic actions rather than restricting all autonomous driving functions.
3Measurement precision
If the travelable region is set on the map, then navigation precision is improved, but the complexity of region mapping and obstacle verification increases
Solution Approach 1:
The map data structure serves multiple functions: it defines the travelable region for navigation, provides a reference frame for obstacle position verification, and enables contradiction detection. This multi-functionality reduces the need for separate verification systems, managing complexity while maintaining precision.
Data Source
Figure 1
Figure 2(a)~2(b)
Figure 3(a)~3(c)
AI summary
A vehicle periphery information verification device (1) comprises: an obstacle detecting unit (20) configured to detect an obstacle; an arranging unit (41) configured to arrange a travelable region in which a vehicle is travelable and the detected obstacle on a map; an action determining unit (43) configured to determine an action of a host vehicle by using information on the arranged obstacle; a judging unit (42) configured to judge whether the arranged obstacle is arranged in the travelable region; and a prohibiting unit (44) configured to prohibit the action determining unit (43) from determining the action of the host vehicle when the obstacle is determined to be not arranged in the travelable region.