Autonomous Grasping On Quadruped Robot With Task Level Interaction

TL;DR

Problem: enabling autonomous grasping on a quadruped platform. Approach: hardware integration of a Lite3 quadruped with OpenManipulator-X, ROS-driven control, and a web-based task-level interface; perception and grasp planning through YOLOv8n and GraspNet. Contributions: a two-subsystem hardware architecture coordinated by a Perception Host; an FSM-based autonomous manipulation pipeline with GraspNet-based grasp selection and a three-stage filter; real-world validation showing 75% success across 12 trials. Significance: demonstrates feasibility of semi-autonomous service robotics using quadruped platforms in real environments.

Abstract

Quadruped robots are increasingly used in various applications due to their high mobility and ability to operate in diverse terrains. However, most available quadruped robots are primarily focused on mobility without object manipulation capabilities. Equipping a quadruped robot with a robotic arm and gripper introduces a challenge in manual control, especially in remote scenarios that require complex commands. This research aims to develop an autonomous grasping system on a quadruped robot using a task-level interaction approach. The system includes hardware integration of a robotic arm and gripper onto the quadruped robot's body, a layered control system designed using ROS, and a web-based interface for human-robot interaction. The robot is capable of autonomously performing tasks such as navigation, object detection, and grasping using GraspNet. Testing was conducted through real-world scenarios to evaluate navigation, object selection and grasping, and user experience. The results show that the robot can perform tasks accurately and consistently, achieving a grasping success rate of 75 % from 12 trials. Therefore, the system demonstrates significant potential in enhancing the capabilities of quadruped robots as service robots in real-world environments.

Figures (9)

• Figure 1: Overview of the proposed Task-Level Interaction system. The operator provides high-level inputs (room designation and visual object selection), which are decomposed into structured task levels. The quadruped robot then autonomously executes the corresponding low-level actions—locomotion, navigation, and manipulation—to retrieve the target object.
• Figure 2: Hardware System Diagram
• Figure 3: The web interface enables users to select target rooms (bottom-left), monitor location and control search actions (center), and view robot status (left). Live feeds from the main and arm-mounted cameras are displayed at the top.
• Figure 4: FSM Flow System
• Figure 5: Mapping Result