Although the benefits of teleoperation are well known, the current implementations require high levels of skill which exclude the average user. One of the difficulties is the delay between sending a command and the movement of the robot which can impact an operator’s ability to safely, and effectively perform tasks. Furthermore, the rejection of commands due to unattainable goals (e.g. outside the robot’s joint limits) can leave the operator confused and frustrated. Advancements in research, such as the use of virtual reality (VR) and inverse kinematic techniques are steadily making robot control more feasible for everyday users, but there is still a lack of understanding of what the robot can or cannot do based on its limits and environment. A benefit of VR over other display methods is the ability to customise the immersive environment that the robot is in. In this paper we take advantage of that benefit to create three visual feedback techniques which give the user information on their commands. A user study was conducted comparing our techniques. The results show that higher levels of visual feedback result in quicker task completion times, lower mental workload and a better user experience.

In the teleoperation of robots, the absence of proprioception means that visual information plays a crucial role. Previous research has investigated methods to offer optimal vantage points to operators during teleoperation, with virtual reality (VR) being proposed as a mechanism to give the operator intuitive control over the viewpoint for improved visibility and interaction. However, the most effective perspective for robot operation and the optimal portrayal of the robot within the virtual environment remain unclear. This paper examines the impact of various visual conditions on users’ efficiency and preference in controlling a simulated robot via VR. We present a user study that compares two operating perspectives and three robot appearances. The findings indicate mixed user preferences and highlight distinct advantages associated with each perspective and appearance combination. We conclude with recommendations on selecting the most beneficial perspective and appearance based on specific application requirements.

Object manipulation techniques in immersive virtual environments are either inaccurate or slow. We present a novel technique, PinNPivot, where pins are used to constrain 1DOF/2DOF/3DOF rotations. It also supports 6DOF manipulation and 3DOF translation. A comparison with three existing techniques shows that PinNPivot is significantly more accurate and faster.