Virtual Reality Based Teleoperation with Adaptive Autonomy for Remote Shared Control of Mobile Manipulators

Purpose and goal

The primary focus of this pre-study is on providing safer, attractive and progressive workplaces as well as increasing efficiency in mining operations, through the use of data-driven approaches in operating mining machinery. A two-fold approach is used to address this objectives: adapting to operators´ behaviors and learning in machine control for smooth and fault safe operations. The project specifically focuses on the use of a general purpose collaborative robotic arms as a generalization over variety of manipulators used in modern mining machinery.

Expected results and effects

The results of the project are expected to contribute to the following areas: 1. Increased safely for machine operators through the ability of the system to be remotely operated from distance with high efficiency, thus eliminating the need for the operator to be in the hazardous areas. 2. Improved working conditions by using adaptive immersive virtual reality interfaces to allow improved performance. 3. Better environmental performance through the use of adaptivity in teleoperated solutions.

Planned approach and implementation

The main focus of this pre-study is to explore and developed machine learning approaches to allow uncertainties in the manipulator model and operator behavior. Two scientific work packages on learning robot model and understanding operator´s behaviors are interconnected with an integration work package to allow rapid testing and demonstration of results. Communication and dissemination is implemented in the form of scientific publications and public outreach.