The role of mobile robots for cleaning and sanitation purposes is increasing worldwide. Disinfection and hygiene are two integral parts of any safe indoor environment, and these factors become more critical in COVID-19-like pandemic situations. Door handles are highly sensitive contact points that are prone to be contamination. Automation of the door-handle cleaning task is not only important for ensuring safety, but also to improve efficiency. This work proposes an AI-enabled framework for automating cleaning tasks through a Human Support Robot (HSR). The overall cleaning process involves mobile base motion, door-handle detection, and control of the HSR manipulator for the completion of the cleaning tasks. The detection part exploits a deep-learning technique to classify the image space, and provides a set of coordinates for the robot. The cooperative control between the spraying and wiping is developed in the Robotic Operating System. The control module uses the information obtained from the detection module to generate a task/operational space for the robot, along with evaluating the desired position to actuate the manipulators. The complete strategy is validated through numerical simulations, and experiments on a Toyota HSR platform.

Mobile robots play a crucial role in cleaning, maintenance, and surveillance applications. This article advocates for the use of a novel robust output feedback based path following controller, for a class of self-reconfigurable mobile robot under actuator saturation. The reconfigurability property of such platforms is captured via an uncertain Euler-Lagrange dynamics. The proposed control framework estimates the unmeasurable states and the uncertain dynamics terms through two extended high gain observers, whereas the actuator limits are honored via a fast dynamic compensator. The closed-loop stability is analyzed via contraction theory, which, compared to the conventional Lyapunov based approaches, avoids the requirement of arbitrarily large controller and observer gains. Such a feature is of particular interest in view of actuator saturation. The experimental results with PANTHERA self-reconfigurable robot validate the effectiveness of the proposed technique over the state of the art.

The cleaning robots are going through some significant development in recent years, driven by the greater market penetration and the demand for better cleaning performance. However, most of the robots have problems covering the total cleaning area given geometric limitations of the platforms in relation to the cleaning stage, given furniture and architecture. In this paper, we describe the hTrihex platform, which is a self-reconfigurable adaptive robot that attains three different configurations. The change in configuration gives rise to variation in the kinematic model, which in turn changes the behavior of robot locomotion. The change in configuration and the robot position is evaluated using the measurements from the on-board sensors like the encoders, LIDAR, and inertial measuring unit (IMU). These data are then utilized to develop a new cascade control strategy for motion control of h-Trihex. The control cascade comprises a robust backstepping controller in the outer loop for tackling the varying kinematic and a conventional PID in the inner loop for speed control. This cascade is appended by a PID controller, for correcting the wheel orientation after each configuration change. The proposed design assures asymptotic path tracking and satisfactory closed-loop performance, which are validated through numerical simulations and experiments.

The pavement inspection task, which mainly includes crack and garbage detection, is essential and carried out frequently. The human-based or dedicated system approach for inspection can be easily carried out by integrating with the pavement sweeping machines. This work proposes a deep learning-based pavement inspection framework for self-reconfigurable robot named Panthera. Semantic segmentation framework SegNet was adopted to segment the pavement region from other objects. Deep Convolutional Neural Network (DCNN) based object detection is used to detect and localize pavement defects and garbage. Furthermore, Mobile Mapping System (MMS) was adopted for the geotagging of the defects. The proposed system was implemented and tested with the Panthera robot having NVIDIA GPU cards. The experimental results showed that the proposed technique identifies the pavement defects and litters or garbage detection with high accuracy. The experimental results on the crack and garbage detection are presented. It is found that the proposed technique is suitable for deployment in real-time for garbage detection and, eventually, sweeping or cleaning tasks.

Reconfigurable robots provide an attractive option for cleaning tasks, thanks to their better area coverage and adaptability to changing environment. However, the ability to change morphology creates drastic changes in the reconfigurable robot dynamics, and existing control design techniques do not take this into account. Neglecting configuration changes can lead to performance degradation and, in the worst scenarios, instability. This paper proposes to embed the changes arising from reconfiguration in the control design, via a switched uncertain Euler-Lagrangian model. Accordingly, a novel switched adaptive design is proposed for trajectory tracking. Closed-loop stability is assured using the multiple Lyapunov function framework, and the design is implemented and validated on a self-reconfigurable pavement cleaning mobile robot (PANTHERA). Note to Practitioners —Self-reconfigurable mobile cleaning robots, which can change their configurations as per the application requirements, are now predominantly used for cleaning and maintenance operations because of their better area coverage, less manpower requirement and consistent performance. However, the state-of-the-art control strategies for conventional robots cannot always ensure stability and performance under the simultaneous effects of configuration changes and uncertainties. The switched Euler-Lagrange model formulated in this work can capture the configuration changes of the robot and the proposed switched adaptive controller can tackle uncertainties of each configurations of the robot. The simulation and experimental results clearly show the potential issues of the state-of-the-art methods and the remarkable benefits of the proposed approach.

Infectious diseases are caused by pathogenic microorganisms, whose transmission can lead to global pandemics like COVID-19. Contact with contaminated surfaces or objects is one of the major channels of spreading infectious diseases among the community. Therefore, the typical contaminable surfaces, such as walls and handrails, should often be cleaned using disinfectants. Nevertheless, safety and efficiency are the major concerns of the utilization of human labor in this process. Thereby, attention has drifted toward developing robotic solutions for the disinfection of contaminable surfaces. A robot intended for disinfecting walls should be capable of following the wall concerned, while maintaining a given distance, to be effective. The ability to operate in an unknown environment while coping with uncertainties is crucial for a wall disinfection robot intended for deployment in public spaces. Therefore, this paper contributes to the state-of-the-art by proposing a novel method of establishing the wall-following behavior for a wall disinfection robot using fuzzy logic. A non-singleton Type 1 Fuzzy Logic System (T1-FLS) and a non-singleton Interval Type 2 Fuzzy Logic System (IT2-FLS) are developed in this regard. The wall-following behavior of the two fuzzy systems was evaluated through simulations by considering heterogeneous wall arrangements. The simulation results validate the real-world applicability of the proposed FLSs for establishing the wall-following behavior for a wall disinfection robot. Furthermore, the statistical outcomes show that the IT2-FLS has significantly superior performance than the T1-FLS in this application.