Research on Indoor Robot Based on Monocular Vision

In this paper, a robot with a single camera is designed for the indoor environment. This robot which equipping Android mobile client contains MCU, camera and related sensors. The user can set the properties of the robot through the Android software. Also, the robot has a total of five different operating modes: obstacle avoidance mode, control mode, edge mode, the tracking mode and monitoring mode. This paper focuses on the realization of obstacle avoidance which combines a single camera with the sensor data. When the object state changes, the object will be identified by the camera. The three-dimensional coordinates of the object are calculated. At the same time, the sensor data is acquired to predict the trajectory of the robot, so as to achieve the tracking of moving objects in the visual field. The experiment results show that the target detection results are accurate, and the success rate is high.