Improving Human-Robot Interaction Utilizing Learning and Intelligence: A Human Factors-Based Approach

Abstract

Several decades of development in the fields of robotics and automation has resulted in human-robot-interaction being commonplace, and the subject of intense study. These interactions are particularly prevalent in manufacturing, where human operators have been employed in a number of robotics and automation tasks. The presence of human operators continues to be a source of uncertainty in such systems, despite the study of human factors, in an attempt to better understand these variations in performance. Concurrent developments in intelligent manufacturing present opportunities for adaptability within robotic control. This work examines relevant human factors and develops a framework for integrating the necessary elements of intelligent control and data processing to provide appropriate adaptability to robotic elements, consequently improving collaborative interaction with human colleagues. A neural network-based learning approach is used to predict the influence on human task performance and use these predictions to make informed changes to programmed behaviour, and a methodology developed to further explore the application of learning techniques to this area. The work is supported by an example case-study, in which a simulation model is used to explore the application of the developed system, and its performance in a real-world production scenario. The simulation results reveal that adaptability can be realised with some relatively simple techniques and models if applied in the right manner and that such adaptability is helpful to tackle the issue of performance disparity in manufacturing operations. NTP: This paper presents research into the application of intelligent methodologies to this problem and builds a framework to describe how this information can be captured, generated and used, within manufacturing production processes. This framework helps identify which areas require further research and serves as a basis for the development of a methodology, by which a control system may enable adaptable behaviour to reduce the impact of human performance variation and improve human-machine-interaction. The paper also presents a simulation-based case study, to support the development and evaluate the presented control system on a representative real-world problem. The methodology makes use of a machine learning approach to identify the complex influence of a number of identified human factors on human performance. This knowledge can be used to adjust the robotic behaviour to match the predicted performance of a number of different operators over a number of scenarios. The adaptability reduces performance disparity, reducing idle times and enabling leaner production through WIP reduction. Future work

Control Techniques for Safe, Ergonomic, and Efficient Human-Robot Collaboration in the Digital Industry: A Survey

“…To test the control system, a 1 DOF testbed power assist robotic system is defined and several position and force control strategies for the system are set and compared, showing the best performance of the method in case of position control. The complex influence of several human factors on the operator performance is also identified by using a neural network based learning approach in the paper [110] by Oliff et al, where a framework that integrates intelligent control and data processing is developed to improve coexistence with human colleagues in the manufacturing context. In another article [111] by Shen et al, an innovative framework is developed to improve HRC by endowing the robot with the capability of understanding human personalities during face-to-face interaction.…”

Towards Designing Adaptive and Personalized Work Systems in Manufacturing

“…This has led to an interconnection between the virtual and the physical world, also known as twinning (Jones 2020). Digital twins and CPPS allow digital processing and planning with closely interconnected physical manipulation to create self-X capabilities, thus self-aware, selfconfigurable, self-optimizing, and self-predictive systems or artefacts (Oliff et al, 2020). The notion of self-X implies the adaptation and alignment of behaviour, state and positions of objects towards desired system output.…”

Explainable Hierarchical Imitation Learning for Robotic Drink Pouring

“…Moreover, model-based approaches for programming the robot for task execution are time-consuming and of lowefficiency. Therefore, it is desirable to employ intelligent methodologies in robotics and automation tasks to enable the quick deployment of service robot to assist humans' daily life [5].…”