Control scheme for human-robot co-manipulation of uncertain, time-varying loads

Abstract

The collaboration between a human operator and a robotic manipulator for performing a joint task is the main topic of this article. One of the most common tasks involves transporting and positioning a load. The proposed control strategy allows the comanipulation of loads with unknown, time-varying mass. The control algorithm is of a cyclic nature, where each cycle is subdivided into two stages. The first phase is the estimation of the human force, since the only connection between the robot and the human is via the load. In the second phase, a scaled version of the estimated human force is applied to the load. The proposed approach is illustrated by means of simulation results for a two-link robot manipulator.

Assisting Operators in Heavy Industrial Tasks: On the Design of an Optimized Cooperative Impedance Fuzzy-Controller With Embedded Safety Rules

“…Human intentions of motion are estimated on the basis of the derivative of the interaction force. In Lichiardopol et al ( 2009 ) a control strategy allowing the co-manipulation of loads with unknown and time-varying mass is proposed. The human applies force is estimated and a scaled version of such estimation is applied to the load to move the manipulated part.…”

Dual Arm Co-Manipulation Architecture with Enhanced Human–Robot Communication for Large Part Manipulation

“…Within the different topics posed in human-robot collaboration, force control is one of the most studied fields, with many approaches and algorithms to take advantage of the force based interaction. Lichiardopol et al [3] propose a control scheme for human-robot co-manipulation with a single robot, where the system estimates an unknown and time-varying mass as well as the force applied by the operator. Dimeas and Aspragathos [5,7] pose a method to detect unstable behavior and stabilize the robot with an online adaptation of the admittance control gains, adding reinforcement learning to estimate parameters for effective cooperation.…”

“…Even so, the successful development of cooperative operations between the operators and robots gives rise to many challenges. Beginning from the low-level robot control [3] and ending with the social and acceptance aspects of these kinds of applications [4], many facets must be tackled during the implementation phase.…”

Estimating robot end-effector force from noisy actuator torque measurements

“…A lot of the existing approaches are based on observers [1], [2], [3], [4], [5] or neural networks [6]. The first force estimation algorithm presented in this paper takes a different approach: it uses a filtered dynamic model, which eliminates the need for acceleration measurements, and then applies recursive least-squares estimation with exponential forgetting to estimate the force.…”