Vision-Guided Multirotor Net Collection on Reflective Water Surfaces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous object collection from water surfaces using UAVs faces challenges such as aerial manipulation, object collection, and landing on a moving object, particularly due to random motion of floating objects and partially submerged objects.
Innovation Solution
A multirotor system equipped with a boundary layer sliding mode control (BLSMC) method, a dynamic sliding manifold, a robust vision-based controller, and an integrated net system for object detection and collection, which includes a computationally efficient contour-based algorithm for specularity removal and a constrained model predictive control (MPC) for optimal sliding surface determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a robotic manipulator is added to the multirotor for aerial grasping, then the grasping capability is improved, but the aircraft gross weight increases and flight time decreases
Solution Approach 1:
The patent extracts the grasping function from a complex robotic manipulator and implements it through a simpler net-based mechanism. The net is deployed from the multirotor to entangle and capture floating objects, eliminating the need for heavy robotic arms while maintaining aerial grasping capability.
Solution Approach 2:
The patent replaces the mechanical robotic manipulator system with a net deployment mechanism. Instead of using motors, joints, and grippers, the system uses a passive net that is released and allows the floating object to become entangled, significantly reducing the mechanical complexity and weight.
2Measurement precision
If the multirotor approaches the floating object for collection, then the collection precision is improved, but the object motion becomes more random due to propeller outwash
Solution Approach 1:
The patent applies preliminary action by deploying the net before the multirotor reaches the floating object. The net is released in advance, creating a capture zone that will entangle the object as it drifts into the net, allowing collection without requiring the multirotor to approach closely.
Solution Approach 2:
The net serves as an intermediary between the multirotor and the floating object. Instead of the multirotor directly contacting or closely approaching the object, the net acts as a mediator that passively captures the object through entanglement as it drifts into the capture zone.
3Difficulty of detecting and measuring
If vision-based tracking is used to monitor floating objects, then the object detection capability is improved, but the tracking reliability decreases due to water surface reflection and glare
Solution Approach 1:
The patent changes the optical parameters by using polarized filters and adjusting camera exposure settings to reduce the impact of water surface reflections and glare. This allows the vision system to maintain reliable tracking despite the challenging optical environment.
Data Source
AI summary
Various embodiments of a vision-guided unmanned aerial vehicle (UAV) system to identify and collect foreign objects from the surface of a body of water are disclosed herein. A vision system and methodology has been developed to reduce reflections and glare from a water surface to better identify an object for removal. A linearized polarization filter and a specularity-removal algorithm is used to eliminate excessive reflection and glare. A contour-based detection algorithm is implemented for detecting the targeted objects on water surface. Further, the system includes a boundary layer sliding mode control (BLSMC) methodology to reduce and minimize position and velocity errors between the UAV and object in the presence of modeling and parameter uncertainties due to variation in a moving water surface.