Drone Relative Position Estimation for Moving-Platform Landing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing aerial vehicle control systems face challenges in reliably controlling drones when the home point is a moving body, such as a ship or helicopter, as they may lose sight of the home point, collide with the moving body during take-off or landing, and experience unstable self-position estimation due to changing conditions.
Innovation Solution
An information processing method that estimates the relative position and attitude of the aerial vehicle with respect to a moving body using image classification technology to segment relevant regions for self-position estimation, and switches estimation methods based on distance from the moving body, using vision information for close proximity and object detection for longer distances to maintain accurate tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the home point is set as a moving point on a moving body, then the aerial vehicle can return to a dynamic target, but the aerial vehicle may lose sight of the home point to return to
Solution Approach 1:
The patent applies dynamics by making the estimation method adaptable to changing conditions. The system dynamically switches between different estimation methods (vision-based for close distances, object detection for far distances) based on the aerial vehicle's distance from the moving body, ensuring reliable tracking throughout the entire flight range.
Solution Approach 2:
The patent changes the parameter of estimation method selection based on distance. When the aerial vehicle is within a threshold distance, vision information is used for estimation; when beyond the threshold, object detection methods are employed, optimizing tracking reliability across different operational phases.
2Adaptability or versatility
If the home point is set as a moving point on a moving body, then the aerial vehicle can perform dynamic operations, but the aerial vehicle may collide with the moving body at the time of take-off or landing
Solution Approach 1:
The patent applies preliminary anti-action by using the estimated relative position and attitude information to proactively adjust the aerial vehicle's flight path and control inputs before collision can occur. This allows the system to compensate for the moving body's motion and maintain safe separation during critical phases like take-off and landing.
Solution Approach 2:
The patent implements feedback by continuously estimating the relative position and attitude of the aerial vehicle with respect to the moving body and using this information to adjust control commands in real-time, preventing collision through closed-loop control.
3Adaptability or versatility
If the home point is set as a moving point on a moving body, then the aerial vehicle can perform dynamic missions, but various problems arise in the control of the aerial vehicle
Solution Approach 1:
The patent applies segmentation by dividing the control problem into distinct phases based on distance. The system uses vision-based estimation for close-range operations and object detection for far-range operations, simplifying the control logic by handling each phase with the most appropriate method rather than attempting a single unified approach.
Data Source
AI summary
An information processing method executed by one processor or executed by a plurality of processors in cooperation, the method includes: an estimation step of estimating a relative position or a relative attitude of an aerial vehicle with respect to a moving body; an acquisition step of acquiring information related to a distance between the moving body and the aerial vehicle; and a switching step of switching an estimation method for estimating the relative position or the relative attitude of the aerial vehicle, based on the information related to the distance.