Optimal selective floor cleaning using deep learning algorithms and reconfigurable robot hTetro

Abstract

Floor cleaning robots are widely used in public places like food courts, hospitals, and malls to perform frequent cleaning tasks. However, frequent cleaning tasks adversely impact the robot’s performance and utilize more cleaning accessories (such as brush, scrubber, and mopping pad). This work proposes a novel selective area cleaning/spot cleaning framework for indoor floor cleaning robots using RGB-D vision sensor-based Closed Circuit Television (CCTV) network, deep learning algorithms, and an optimal complete waypoints path planning method. In this scheme, the robot will clean only dirty areas instead of the whole region. The selective area cleaning/spot cleaning region is identified based on the combination of two strategies: tracing the human traffic patterns and detecting stains and trash on the floor. Here, a deep Simple Online and Real-time Tracking (SORT) human tracking algorithm was used to trace the high human traffic region and Single Shot Detector (SSD) MobileNet object detection framework for detecting the dirty region. Further, optimal shortest waypoint coverage path planning using evolutionary-based optimization was incorporated to traverse the robot efficiently to the designated selective area cleaning/spot cleaning regions. The experimental results show that the SSD MobileNet algorithm scored 90% accuracy for stain and trash detection on the floor. Further, compared to conventional methods, the evolutionary-based optimization path planning scheme reduces 15% percent of navigation time and 10% percent of energy consumption.

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

“…Moreover, Table 4 lists the multimodal navigation. The applications of the multimodal navigation consist of entertainment [ 143 ], security [ 133 ], transport [ 2 , 106 , 107 , 110 , 112 , 147 ], assistance [ 2 , 16 , 108 ], exploration [ 3 , 134 , 135 , 138 , 141 ], social applications [ 12 , 140 ], tracking [ 105 , 125 , 131 , 145 , 146 ], caring and monitoring [ 113 ], disaster monitoring or search and rescue [ 136 , 149 ], floor cleaning [ 137 ], wheeled robot [ 109 ], person following [ 129 ], and false-ceiling inspection [ 157 ]. The combination of virtual sensors and neural networks is most commonly used in multimodal navigation, which represents 77.19% of the cited research.…”

“…Ramalingam et al [ 137 ] presented a selective area-cleaning/spot-cleaning framework based on an RGB-D vision sensor and a deep learning algorithm for indoor floor-cleaning robots, as shown in Figure 16 . The human traffic region was traced by a simple online and real-time tracking algorithm and the dirty region was detected by a single-shot detector MobileNet.…”

Vision-based dirt distribution mapping using deep learning

“…However, this sort of concept has yet to be fully realized, and merely a few supporting concepts can be found in the literature. For example, Ramalingam et al 16 proposed a closed-circuit television (CCTV)-based system to guide a robot for selective cleaning. Here, the dirt location and human activities are detected, and an optimum path will be created for the robot for spot cleaning resulting in higher efficiency.…”