Aerial Grasping via Maximizing Delta-Arm Workspace Utilization

TL;DR

This paper presents a novel planning framework for aerial grasping that maximizes workspace utilization by optimizing trajectories and employing neural networks to handle complex kinematics, validated through simulations and real-world tests.

Contribution

It introduces a new optimization-based planning method that maximizes workspace utilization for aerial manipulators using neural networks to handle non-convex constraints.

Findings

Improved workspace utilization in aerial grasping tasks.

Effective neural network models for workspace feasibility and kinematics.

Successful validation in both simulation and real-world experiments.

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.