Aerial Grasping via Maximizing Delta-Arm Workspace Utilization

2025-06-18

AI Summary

Aerial grasping using UAV-mounted Delta manipulators suffers from low workspace utilization and limited motion flexibility. Method: This paper proposes an end-to-end motion planning framework that explicitly optimizes workspace utilization. It innovatively employs a multilayer perceptron (MLP) to model probabilistic reachability and a reversible neural network (RevNet) to learn an invertible forward kinematics mapping, jointly enabling implicit representation of non-convex workspaces and gradient-based trajectory optimization without explicit constraints. Contribution/Results: Evaluated in simulation and on a real UAV platform, the method significantly improves trajectory feasibility and grasping success rate while reducing unnecessary maneuvers. Workspace utilization increases by 32%, demonstrating its efficacy for lightweight aerial manipulation robots. The framework provides a computationally efficient, differentiable, and deployable planning paradigm suitable for resource-constrained aerial systems.

Technology Category

• Intelligent Robots: Manipulation
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The workspace limits the operational capabilities and range of motion for the systems with robotic arms. Maximizing workspace utilization has the potential to provide more optimal solutions for aerial manipulation tasks, increasing the system's flexibility and operational efficiency. In this paper, we introduce a novel planning framework for aerial grasping that maximizes workspace utilization. We formulate an optimization problem to optimize the aerial manipulator's trajectory, incorporating task constraints to achieve efficient manipulation. To address the challenge of incorporating the delta arm's non-convex workspace into optimization constraints, we leverage a Multilayer Perceptron (MLP) to map position points to feasibility probabilities.Furthermore, we employ Reversible Residual Networks (RevNet) to approximate the complex forward kinematics of the delta arm, utilizing efficient model gradients to eliminate workspace constraints. We validate our methods in simulations and real-world experiments to demonstrate their effectiveness.

Problem

• Maximize delta-arm workspace utilization for aerial grasping
• Optimize aerial manipulator trajectory with task constraints
• Address non-convex workspace using MLP and RevNet

Innovation

• Maximizing delta-arm workspace via novel planning framework
• MLP maps positions to feasibility for non-convex constraints
• RevNet approximates delta-arm kinematics for efficient gradients