A Synergistic Approach to the Real-Time Estimation of the Feet Ground Reaction Forces and Centers of Pressure in Humans With Application to Human–Robot Collaboration

Abstract

This letter proposes a novel technique for the realtime estimation of the human feet ground reaction forces (GRFs) and centers of pressure (CoP) from the whole-body CoP and body configuration. The estimated variables are used for the estimation of the overloading torques in human joints during double support, with the aim to provide an evaluation of the human ergonomics while performing heavy manipulation tasks with a robot, or when exposed to an external force. The estimation of the feet GRFs and CoP is achieved using a synergistic approach that combines a simplified geometrical model of the whole-body CoP and a learning technique to address the underdetermined force distribution problem. First, a statically equivalent serial chain (SESC) model which enables the whole-body CoP estimation is identified. Then, the estimated whole-body CoP and the simplified body pose information are used for the training and validation of the learning technique. The displacements between the feet CoP estimated using the synergistic model and the values measured using insole sensors, along with the GRFs vectors are finally used for the estimation of the body overloading joint torques. The proposed synergistic model is first validated experimentally in five subjects. Next, its real-time efficacy is assessed in a human-robot load sharing task, in which the robot trajectories are optimized to minimize the effect of the external load on the human joints. Index Terms-Human-centered robotics, human factors and human-in-the-loop, physical human-robot interaction. I. INTRODUCTION I N THE last few years, the great potential and the countless benefits of human-robot collaboration (HRC) are becoming increasingly evident among the scientific and industrial communities. The robots can reduce fatigue and stress by Manuscript

Human-robot collaboration and machine learning: a systematic review of recent research

“…The system performs with higher safety, compliance and precision (R1) [27] The human assembles a box…”

“…Finally, another subset of papers aimed at reducing the human physical workload during task completion. A common aspect between these papers is the use of biomechanical variables that were minimised using the robotic system, to validate the efficacy of the proposed solution [27]. Another portion of them also looks at the reduction of the interaction forces between robot and user [17,22].…”

“…These cognitive variables, mostly inherited from human biomechanics, are used as quantities to minimise by the control routine following the machine learning algorithm. In [27], the ground reaction force exerted on the user, along with his/her centre of pressure, are used by the robot to find the optimal way of leveraging the effort during a lifting task. In [36], through measurement of the muscular activity of the user, the human fatigue is modelled and used as parameter of interest to regulate the collaboration.…”

A New Overloading Fatigue Model for Ergonomic Risk Assessment with Application to Human-Robot Collaboration

“…Accordingly, the optimisation procedure had to occur multiple times throughout the task, and to focus selectively on the fatigued joint. As a support tool to the optimisation procedure, the subject was provided with a visual feedback we developed in a previous work [28], showing the current body configuration and the current level of the overloading fatigue on the main joints. Fig.5a illustrates the four key moment of the experiment (i.e.…”

Towards Ergonomic Control of Collaborative Effort in Multi-human Mobile-robot Teams

“…Since the matrices M, C, and G are the same in the two conditions, they can be cancelled, lightening the computations. The overloading joint torques considering the force distribution ratios (0 ≤ η j , ζ i ≤ 1 [21]) can then be defined as…”