A Small Form Factor Aerial Research Vehicle for Pick-and-Place Tasks with Onboard Real-Time Object Detection and Visual Odometry

TL;DR

This work presents a small-form-factor UAV platform for onboard pick-and-place tasks in constrained environments, combining a collision-tolerant cage, the GREP gripper, and entirely onboard perception and state estimation (DOPE for 6-DOF object pose and stereo VO). Implemented within a ROS-based Aerial Autonomy framework, the system fuses VO with an EKF and uses a fault-tolerant state machine to robustly execute pick-and-place via visual servoing to guide the gripper. Hardware designs and software are open-source, and high-fidelity Gazebo SITL simulation supports rapid development and testing. Experimental validation across 70 hardware trials yields 93% pick success and 86% place success, demonstrating reliable grasping and placement on small UAVs in cluttered and occluded environments, with potential impact on indoor robotics and autonomous manipulation in constrained spaces.

Abstract

This paper introduces a novel, small form-factor, aerial vehicle research platform for agile object detection, classification, tracking, and interaction tasks. General-purpose hardware components were designed to augment a given aerial vehicle and enable it to perform safe and reliable grasping. These components include a custom collision tolerant cage and low-cost Gripper Extension Package, which we call GREP, for object grasping. Small vehicles enable applications in highly constrained environments, but are often limited by computational resources. This work evaluates the challenges of pick-and-place tasks, with entirely onboard computation of object pose and visual odometry based state estimation on a small platform, and demonstrates experiments with enough accuracy to reliably grasp objects. In a total of 70 trials across challenging cases such as cluttered environments, obstructed targets, and multiple instances of the same target, we demonstrated successfully grasping the target in 93% of trials. Both the hardware component designs and software framework are released as open-source, since our intention is to enable easy reproduction and application on a wide range of small vehicles.

Figures (11)

• Figure 1: Aerial research platform, target objects (toy cans and bottles), and view of open gripper, $9.5$ cm across, around $6.5$ cm diameter can.
• Figure 2: Autonomous system in flight grasping a can from a cluttered scene.
• Figure 3: CAD model of vehicle with integrated modular components.
• Figure 4: GREP: Fixed arm with angular motion gripper.
• Figure 5: Software flow diagram showing a simplified pick-and-place state machine, begins at "Waiting to Pick" state.