Design of a low-cost and lightweight 6 DoF bimanual arm for dynamic and contact-rich manipulation

2025-02-24

AI Summary

Current robotic systems struggle with dynamic, contact-intensive manipulation tasks—such as hammering and throwing—due to high inertia, low mechanical compliance, and reliance on expensive torque sensors. This paper introduces ARMADA: a low-cost ($6,100), lightweight, low-inertia, open-source dual-arm 6-DOF robotic platform. Our method integrates high-speed motion (6.16 m/s—twice that of typical collaborative robots), high backdrivability, strong dynamic responsiveness, and zero-cost joint torque estimation within a unified mechatronic and control architecture. It enables zero-shot sim-to-real reinforcement learning policy transfer (using PyBullet + PPO) and real-time human motion mapping. The design leverages 3D-printed structures, low-inertia rear-drive motors, and real-time closed-loop dynamic control. Experimental evaluation demonstrates a 2.5 kg payload capacity and successful execution of non-prehensile manipulation, bimanual object tossing, and dynamic motion reproduction.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Adversarial Learning & Robustness
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Dynamic and contact-rich object manipulation, such as striking, snatching, or hammering, remains challenging for robotic systems due to hardware limitations. Most existing robots are constrained by high-inertia design, limited compliance, and reliance on expensive torque sensors. To address this, we introduce ARMADA (Affordable Robot for Manipulation and Dynamic Actions), a 6 degrees-of-freedom bimanual robot designed for dynamic manipulation research. ARMADA combines low-inertia, back-drivable actuators with a lightweight design, using readily available components and 3D-printed links for ease of assembly in research labs. The entire system, including both arms, is built for just $6,100. Each arm achieves speeds up to 6.16m/s, almost twice that of most collaborative robots, with a comparable payload of 2.5kg. We demonstrate ARMADA can perform dynamic manipulation like snatching, hammering, and bimanual throwing in real-world environments. We also showcase its effectiveness in reinforcement learning (RL) by training a non-prehensile manipulation policy in simulation and transferring it zero-shot to the real world, as well as human motion shadowing for dynamic bimanual object throwing. ARMADA is fully open-sourced with detailed assembly instructions, CAD models, URDFs, simulation, and learning codes. We highly recommend viewing the supplementary video at https://sites.google.com/view/im2-humanoid-arm.

Problem

• Low-cost 6 DoF bimanual arm design
• Dynamic and contact-rich manipulation challenge
• Reinforcement learning for real-world application

Innovation

• Low-cost 6 DoF bimanual arm
• Lightweight design with 3D-printed links
• Open-sourced for dynamic manipulation research