An sEMG-Based Human–Robot Interface for Robotic Hands Using Machine Learning and Synergies

Abstract

Developing natural control strategies represents an intriguing challenge in the design of Human-Robot Interface (HRI) systems. The teleoperation of robotic grasping devices, especially in industrial, rescue and aerospace applications, is mostly based on non-intuitive approaches, such as remote controllers. On the other hand, recent research efforts target solutions that mimic the human ability to manage multi-finger grasps and finely modulate grasp impedance. Since electromyography (EMG) contains information about human motion control, it is possible to leverage such neuromuscular knowledge to teleoperate robotic hands for grasping tasks. In this article we present a HRI system based on 8 fully-differential EMG sensors connected to a wearable sensor node for acquisition and processing.By virtue of a novel bio-inspired approach, the embedded myocontroller merges pattern recognition and factorization techniques to combine a natural selection of the robotic hand configuration with the proportional control of the related grasps. The HRI system has been fully designed, implemented and tested on two robotic hands: a dexterous anthropomorphic hand and a three-fingered industrial gripper mounted on a robotic manipulator. Results of the test performed on 4 able-bodied subjects show success rates greater than 90% reached in grasping objects that require different hand shapes and impedance regulations for the task completion. The outcomes also show that the users modulate the bio-inspired degrees of control in a natural manner, proving the pertinence of the proposed system for an effective human-like control of robotic grasping devices in a wearable form-factor.

Exploring Arm Posture and Temporal Variability in Myoelectric Hand Gesture Recognition

“…The Electromyogram (EMG) is the biopotential signal resulting from muscular activity, it can be sensed using noninvasive surface electrodes and processed to implement myoelectric Human-Machine Interfaces (HMIs). Recent research highlights its potential to enable new interaction paradigms in applications such as natural prosthetic control [1], robot interaction [2], game or mobile interfaces [3]. Hand gesture recognition based on forearm EMG signals is an enabling technology for the development of advanced and intuitive interaction strategies [4].…”

A Control Architecture for Grasp Strength Regulation in Myocontrolled Robotic Hands Using Vibrotactile Feedback: Preliminary Results

“…Note that the control by the user of different robotic grasp actions at the same time was not within the interests of the present work. In this relation, NMF has been successfully used for semi-unsupervised, human-like myocontrol based on muscle synergies related to user's opening/closing hand motions [20]. For the UB Hand controller, the grasp was realized using a myocontroldriven input signal for the regulation of the robotic hand joint velocities, combined with a postural synergy based approach [21].…”

“…Then the NMF algorithm was applied for the computation of the muscular synergy matrix S M ∈ R n E ×n U , according to the equation E of f line ≈ S M U of f line , with U ∈ R n U ×n the socalled offline neural drives matrix, and n U the number of supraspinal neural drives (in our case n U = 2). For further details on the application of NMF for the estimation of the muscular synergy matrix during hand closure motions, refer to our previous work [20].…”

Online Learning and Classification of EMG-Based Gestures on a Parallel Ultra-Low Power Platform Using Hyperdimensional Computing

“…Such method is highly reliable and amenable to implement on a wearable system. However, it requires a long learning curve and high levels of concentration of the user, since it is a non-intuitive interface [5].…”