Autonomous Visual Navigation System Based on a Single Camera for Floor-Sweeping Robot

Abstract

The indoor sweeping robot industry has developed rapidly in recent years. The current sweeping robot environment perception sensor configuration is more diverse and generally does not have active garbage detection capabilities. Advances in computer vision technology, artificial intelligence, and cloud computing technology have provided new possibilities for the development of sweeping robot technology. This paper conceptualizes a new autonomous visual navigation system based on a single-camera sensor for floor-sweeping robots. It investigates critical technologies such as floor litter recognition, environmental perception, and dynamic local path planning based on depth maps. The system is applied to the TurtleBot robot for experiments, and the results show that the mAP accuracy of the autonomous visual navigation system for fine trash recognition is 91.28%; it reduces the average relative error of depth perception by 10.4% compared to conventional methods. Moreover, it has greatly improved the dynamics and immediacy of path planning.

Bioinspired Perception and Navigation of Service Robots in Indoor Environments: A Review

“…More precisely, Table 1 lists the vision-based approaches with their contributions, sensors, and real-time performance, and the most popular method is the optic flow, which is used by bees. The vision-based applications include transport [ 9 , 18 , 39 , 46 , 50 , 60 ], exploration [ 8 , 19 , 24 , 26 , 40 , 41 , 45 , 49 ], tracking and assist [ 31 , 56 , 58 ], security and surveillance [ 7 , 52 ], homing and searching [ 5 , 54 , 55 ], floor cleaning [ 6 , 43 ], and search and rescue [ 44 ]. Only 43.75% of the cited papers indicate they can operate in real time, which is achieved by processing or computation speed, minimal computation load, or predefined parameters.…”

“…A camera-based autonomous navigation system was conceptualized for floor-sweeping robots in [ 6 ], including inspection, targeting, environment perception, local path planning, and directional calibration, as demonstrated in Figure 8 . It achieved image processing and map planning by a superpixel image segment algorithm, but it could interfere with light.…”

“…With animal perception, cognition, body architecture, and behavior research, biorobotics and neurorobotics bring attention to robotics research. Bioinspired robots can perform several tasks, including transport [ 5 ], floor-sweeping [ 6 ], security [ 7 ], caring [ 2 ], and exploration [ 8 ].…”