An Approach to Acquire Path-Following Skills by Industrial Robots From Human Demonstration

Angel Rodriguez-Linan,Alan Maldonado-Ramirez,Antonio Zalapa-Elias,Jose Luis Navarro-Gonzalez,Luis Torres-Trevino,Ismael Lopez-Juarez,Pamela Chinas-Sanchez

doi:10.1109/access.2021.3086701

Abstract

Industrial robots have mainly been programmed by operators using teach pendants in a point-to-point scheme with limited sensing capabilities. New developments in robotics have attracted a lot of attention to robot motor skill learning via human interaction using Learning from Demonstration (LfD) techniques. Robot skill acquisition using LfD techniques is characterised by a high-level stage in charge of learning connected actions and a low-level stage concerned with motor coordination and reproduction of an observed path. In this paper, we present an approach to acquire a path-following skill by a robot in the low-level stage which deals with the correspondence of mapping links and joints from a human operator to a robot so that the robot can actually follow a path. We present the design of an Inertial Measurement Unit (IMU) device that is primarily used as an input to acquire the robot skill. The approach is validated using a motion capture system as ground truth to assess the spatial deviation from the human-taught path to the robot's final trajectory.

Highlights

Robots have been used in the automotive, electrical, metal, chemical and food industry some other areas have began using service robots in everyday tasks such as vacuuming, sweeping or grass mowing
The application of the Adaptive Kalman Filter (AKF) filter resulted in lower errors in both the static and the dynamic case, so this filter was implemented in the Inertial Measurement Unit (IMU) device during tests
WORK Robot skill acquisition using Learning from Demonstration (LfD) techniques involves a high-level stage in charge of learning connected actions, whereas a low-level stage is concerned with motor coordination to reproduce and imitate an observed human path

Summary

INTRODUCTION

Robots have been used in the automotive, electrical, metal, chemical and food industry some other areas have began using service robots in everyday tasks such as vacuuming, sweeping or grass mowing. ORIGINAL CONTRIBUTION The human arm does not usually remain in a completely static state, as there are small movements (such as breathing) that must be discarded as a movement event in LfD applications in robots In this sense, our original contribution is twofold since it uses a novel fusion algorithm based on the variance value for the motion detection to detect static states and it considers the development of a flexible portable device that is not restricted to structured environments. Considering these conditions, this approach could be more attractive in several other applications, as it has recently been reported in [1], [29]–[32] This IMU-based platform for motion capture can be embedded in high-level stages of LfD operations where the robot is directed by the human operator to the desired positions by the movements of his body (postures).

ESTIMATION OF POSITION AND ORIENTATION OF THE HUMAN HAND BY THE IMU DEVICE

ORIENTATION ESTIMATION

POSITION ESTIMATION

CONCLUSION AND FUTURE WORK