Model-based sensorless robot collision detection under model uncertainties with a fast dynamics identification

Abstract

This article presents a novel model-based sensorless collision detection scheme for human-robot interaction. In order to recognize external impacts exerted on the manipulator with sensitivity and robustness without additional exteroceptive sensors, the method based on torque residual, which is the difference between nominal and actual torque, is adopted using only motor-side information. In contrast to classic dynamics identification procedure which requires complicated symbolic derivation, a sequential dynamics identification was proposed by decomposing robot dynamics into gravity and friction item, which is simple in symbolic expression and easy to identify with least squares method, and the remaining structure-complex torque effect. Subsequently, the remaining torque effect was reformulated to overcome the structural complexity of original expression and experimentally recovered using a machine learning approach named Lasso while keeping the involving candidates number reduced to a certain degree. Moreover, a state-dependent dynamic threshold was developed to handle the abnormal peaks in residual due to model uncertainties. The effectiveness of the proposed method was experimentally validated on a conventional industrial manipulator, which illustrates the feasibility and simplicity of the collision detection method.

A Mixed-Perception Approach for Safe Human–Robot Collaboration in Industrial Automation

“…Towards this goal, it is imperative firstly, to detect the human-robot contact and secondly, distinguish between intentional and incidental contacts, a process called collision detection. Some researchers propose sensor-less procedures for detecting a collision based on the robot dynamics model [52,53], but through momentum observers [52,[54][55][56][57], using extended state observers [58], vibration analysis models [59], finite-time disturbance observers [56], energy observers [57], or joint velocity observers [60]. Among these methods, the momentum observer is the most common method of collision detection, because it provides better performance compared to the other methods, although the disadvantage is that it requires precise dynamic parameters of the robot [61].…”

Disturbance Recognition and Collision Detection of Manipulator Based on Momentum Observer

“…It is difficult to extract the friction from the joint torque. A reliable method for the above problem was proposed by the authors of [ 37 ]. Considering that the speed direction is independent of the calculated joint torque by dynamics model, the two types of trajectories are characterized by the same position, acceleration, and speed value, and the opposite speed direction.…”

A Framework for Inclusion of Unmodelled Contact Tasks Dynamics in Industrial Robotics

“…As a result, they are now offering different Industry 4.0 functionalities, connectivity, and simulation-related options, at least for their new and upcoming products [ 3 ]. Furthermore, related topics that have made undisputedly valuable contributions to novel industrial concepts are: collision detection and avoidance [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 ], physical human–robot interaction [ 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 ], kinesthetic guidance [ 19 , 20 , 26 ], motion planning, digital twins, augmented reality, and facilitated programming.…”