Construction of a Multiple-DOF Underactuated Gripper with Force-Sensing via Deep Learning

2024-07-15

Robotics: Science and Systems

AI Summary

To address the challenge of implementing sensorless force-feedback control in underactuated grippers, this work proposes a novel single-actuator, dual-three-joint, five-link underactuated gripper. It features a stacked dual-four-bar mechanism enabling autonomous switching between parallel and enveloping grasping modes. We introduce the first LSTM-based time-series inversion method that estimates contact force from motor current measurements—eliminating the need for physical force sensors. A coupled-decoupled hybrid kinematic and dynamic model is established to precisely relate motor current, joint pose, and contact force. Experiments demonstrate a maximum payload capacity of 1.2 kg, contact force estimation error below 8.3%, and stable grasping of objects with diameters ranging from 10 to 120 mm. This work makes original contributions in three aspects: (i) innovative mechanical design, (ii) sensorless force perception, and (iii) closed-loop control integration—significantly enhancing the versatility, robustness, and engineering practicality of underactuated grippers.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

We present a novel under-actuated gripper with two 3-joint fingers, which realizes force feedback control by the deep learning technique- Long Short-Term Memory (LSTM) model, without any force sensor. First, a five-linkage mechanism stacked by double four-linkages is designed as a finger to automatically achieve the transformation between parallel and enveloping grasping modes. This enables the creation of a low-cost under-actuated gripper comprising a single actuator and two 3-phalange fingers. Second, we devise theoretical models of kinematics and power transmission based on the proposed gripper, accurately obtaining fingertip positions and contact forces. Through coupling and decoupling of five-linkage mechanisms, the proposed gripper offers the expected capabilities of grasping payload/force/stability and objects with large dimension ranges. Third, to realize the force control, an LSTM model is proposed to determine the grasping mode for synthesizing force-feedback control policies that exploit contact sensing after outlining the uncertainty of currents using a statistical method. Finally, a series of experiments are implemented to measure quantitative indicators, such as the payload, grasping force, force sensing, grasping stability and the dimension ranges of objects to be grasped. Additionally, the grasping performance of the proposed gripper is verified experimentally to guarantee the high versatility and robustness of the proposed gripper.

Problem

• Design a low-cost under-actuated gripper with force feedback
• Develop kinematics models for accurate fingertip force sensing
• Implement LSTM-based force control without physical sensors

Innovation

• Under-actuated gripper with double four-linkage fingers
• LSTM model for force feedback without sensors
• Kinematics and power transmission theoretical models