Evaluation of the effects of user profile and interface characteristics on performance during robotic teleoperation.

Abstract

In the mining industry, operator safety has always been a crucial point, and therefore demands attention. In addition, given the high investments made in the enterprise, the search for a consistent and e cient production process is also relevant. To increase productivity and safety, teleoperation techniques can be used to allow the operator to be removed from the risk areas and operate the equipment from a safe distance. The work presented in this master's thesis was developed at the Instituto Tecnológico Vale (ITV) as a part of a broader project that aims the development of a conceptual framework that enables the analysis and validation of the application of teleoperation techniques in mining equipment. These equipment could be either used for exploration, engineering or extraction. More speci cally, this framework will permit the analysis of relevant information for teleoperation. This work represents the initial step taken, which comprises the execution of proof-of-concept tests in laboratory using a robotic system. Di erent teleoperation interfaces schemes consisting of video and haptic devices are analyzed under di erent aspects and the user point of view is taken into account. The goal is to analyze the e ects of the force feedback and of di erent visual feedback during teleoperation and to evaluate which prede ned users characteristics can contribute to better performance during teleoperation. Experimental tests were performed using a commercial mobile robotic platform Seekur Jr and teleoperation interfaces consisting of the haptic devices Novint Falcon and Geomagic Touch. This platform can be programmed in such a way that it operates emulating a mining equipment. For practical application-oriented tests, di erent operations from a mining process can be used as application of study (e.g. explosive charging or excavators operations). For the purposes of this work, interfaces for the control of the robotic platform locomotion were analyzed. In order to evaluate the in uence of interface characteristics and users pro le on performance during teleoperation two experiments were conducted. The experiments consisted of guiding the robotic platform in a scenario containing obstacles, using the di erent interfaces. In this work, delays in communication during teleoperation were not addressed. Results show that the use of force feedback e ectively contributes for better performances in term of task execution time, number of collisions with obstacles and task completion in teleoperation activities. Moreover, additional visual information, such as the environment map, can also increase perception of obstacles.