Home /Research /Real-time human-in-the-loop remote control for a life-size traffic police robot with multiple augmented reality aided display terminals
HRI

Real-time human-in-the-loop remote control for a life-size traffic police robot with multiple augmented reality aided display terminals

Liang Gong, Changyang Gong, Zhao Ma, Lujie Zhao, Zhenyu Wang, Xudong Li, Xiaolong Jing, Haozhe Yang, Chengliang Liu

Year
2017
Citations
22

Abstract

Policing of road traffic is listed in the most hazardous tasks since many traffic police personnel injured in a series of accidents at intersections. A life-size traffic cop robot called “IWI” is invented as an alternative to human traffic police in the street for directing traffic on point duty, and this paper proposes a human-robot cooperation scheme with wearable augmented glasses as an HMI to obtain a more immersing teleoperation experience with the aid of wearable augmented reality glasses. First, a humanoid robot is tailored for mobile surveillance with raspberry pi camera based eyes and omnidirectional Mecanum wheels. Second, the real-time video stream acquired by the on-site robot is distributed to multiple terminals, such as the central sever, the wearable augmented reality glasses and the mobile control tablet. Third, with a depth of focus estimation the augmented indications are displayed to aid to understand the remote scenario. Finally, the human policing results, such as STOP, PULL OVER, TURN-LEFT, are compiled and programmed as simplified patterns to control the robot body/hand pantomime. Experimental results show that the proposed methodology and control scheme is feasible in real-time application with high real-time performance of less than 0.5s latency, and open possibilities of easing the traffic jam via simultaneously scheduling multiple traffic cop robots.

Keywords

Augmented realityHuman-in-the-loopComputer scienceRobotLoop (graph theory)Control (management)Real-time computingHuman–computer interactionSimulationArtificial intelligence

Related papers

Browse all HRI papers