Vehicle Trajectory Correction Using Obstacle Side Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current travel control systems lack the accuracy needed to precisely position a moving body with respect to obstacles when navigating through narrow roads, particularly due to insufficient recognition of the obstacle's side portions and their impact on the vehicle's trajectory.
Innovation Solution
A control apparatus and method that includes a movement planning unit, a correction unit, and a movement control unit, which utilize image recognition data from cameras to continuously correct the vehicle's trajectory based on the recognition of obstacle side portions, ensuring a target margin is maintained to avoid contact with obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional travel control systems are used, then the system complexity is kept simple, but the positioning accuracy with respect to obstacles is insufficient
Solution Approach 1:
The patent implements feedback control by continuously acquiring recognition results of obstacle side portions during vehicle advancement and using these results to calculate correction amounts that are fed back to adjust the movement route. This closed-loop feedback mechanism enables high-accuracy positioning while maintaining manageable system complexity through iterative refinement rather than complex upfront planning.
Solution Approach 2:
The patent performs preliminary recognition of obstacle side portions before final route execution. By acquiring recognition results in advance and calculating correction amounts beforehand, the system prepares positioning adjustments that are then applied during movement, enabling accurate positioning without requiring overly complex real-time processing.
2Measurement precision
If continuous correction based on obstacle recognition is implemented, then positioning accuracy is improved, but the calculation load and processing time increase
Solution Approach 1:
The patent extracts only the necessary information for correction by focusing specifically on recognizing obstacle side portions rather than processing complete obstacle models. This selective extraction of relevant recognition results reduces calculation load while maintaining positioning accuracy, as only the critical side portion data needed for route correction is processed.
Solution Approach 2:
The patent applies partial correction amounts based on recognition results rather than complete recalculation of the entire movement route. By making incremental adjustments to the route based on side portion recognition, the system achieves continuous positioning accuracy improvement without the computational burden of full route recalculation at each step.
3Reliability
If recognition of obstacle side portions is performed, then the ability to avoid contact with protruding portions is improved, but the difficulty of detecting and measuring increases
Solution Approach 1:
The patent applies local quality by focusing recognition efforts specifically on the side portions of obstacles that are relevant to contact avoidance, rather than attempting to recognize and measure all aspects of obstacles. This localized recognition approach improves contact avoidance capability by concentrating measurement resources on the critical areas that affect vehicle-obstacle interaction.
Solution Approach 2:
The patent addresses the detection difficulty by transitioning from traditional two-dimensional image recognition to three-dimensional spatial understanding of obstacle side portions. By calculating correction amounts based on the spatial relationship between the vehicle and obstacle side portions in three-dimensional space, the system improves contact avoidance while managing the complexity of detection through dimensional transformation.
Data Source
AI summary
A control apparatus plans a movement route of a moving body based on a recognition result of an exterior environment of the moving body; corrects the planned movement route, based on a recognition result of an obstacle in the exterior environment of the moving body; and controls the movement of the moving body based on the corrected movement route. The recognition result of the obstacle includes a recognition result of a side portion of the obstacle with respect to the movement route of the moving body, and in the correction, the planned movement route is corrected based on a correction amount continuously obtained based on the recognition result of the obstacle associated with advancement of the moving body.