Contrary to submarine control strategies, pilots of Remotely Operated Vehicles (ROVs) remain remote to the vehicle while manoeuvring and manipulating them through their surrounding environments using tele-operation technology and visual feedback interface. However, due to the lack of three-dimensional (3D) information and limitations with quality and the field vision of the cameras, ROV pilots are not always able to observe and identify external hazards purely through visual feedback, especially when ROVs operate in environments littered with obstacles, which can thus result in collisions. To avoid such occurrences, a haptic controller including a tactile user interface was developed for ROVs in order to enhance the operators' awareness of the working environment and thus improve the performances of the vehicles within such environments. The haptic control technique developed in this project was initially examined within a simulated environment and then applied to control a low-cost open source hardware-based ROV to validate its effectiveness. In the simulation phase, the haptic control interface hardware and software, including the haptic joystick and a novel Artificial Potential Field (APF) technique was developed to assist pilots to safely manoeuvre the simulated ROV through the surrounding environment containing both static and moving obstacles. A robust adaptive control algorithm for the haptic controller was also developed to improve the performance of the haptic control system while maintaining its stability. The proposed technique was then experimentally validated by applying the haptic controller to an observation class ROV, designed and developed for this project using open source hardware and low-cost equipment. To ensure the quality of the ROV's haptic drive system, a multi-layer Kalman filter and advanced control algorithms, such as adaptive PID and robust model-based algorithms, were designed to estimate the vehicle's states and to control its surge, yaw, and heave motions based on a host-target control structure. The experimental results show that the host-target control structure was effectively employed to collect data and control the open source hardware-based ROV in real time. Additionally, the host-target structure was able to overcome the limitation associated with the computational power of the microcontroller, allowing the programmers to develop complex algorithms to process the raw data from low-cost sensors and deal with the nonlinear characteristics of the vehicles. The ROV performances observed from both the simulations and the experimental work indicate that the multi-layer Kalman filter and the adaptive PID algorithms provide acceptable state estimation for the ROV and thus assist the pilot in adequately manoeuvring the vehicle. The simulation results show that the proposed APF technique has the ability to model the potential risk presented by both stationary and moving obstacles. The information is fed back and used as reference signals for the force controller incorporated within the haptic joystick system to generate a haptic force, which allows the ROV pilots to 'feel' the interaction with the surrounding environment. Finally, the simulation and experimental results show that adverse effects, such as parasitic forces and instability caused by model uncertainties and time delays, were effectively mitigated by the robust adaptive force control algorithm. The results and the findings show that haptic technology developed within this project is suitable to assist ROV pilots to safely control the vehicle within hazardous environments.
Copyright 2016 the author Chapter 2, part B appears to be the equivalent of an article published as: Le, K. D., Nguyen, H. D., Ranmuthugala, D., 2014. Design, modelling and simulation of a remotely operated vehicle - Part 2, Journal of computer science and cybernetics, 30(2), 106-116 Copyright 2014 Journal of computer science and cybernetics Chapter 3, part A appears to be the equivalent of an article published as: Le, K. D., Nguyen, H. D., Ranmuthugala, D., Forrest, A., 2015. A heading observer for ROVs under roll and pitch oscillations and acceleration disturbances using low-cost sensors, Ocean engineering, 110(Part A), 152-162 Chapter 4, part B appears to be the equivalent of an article published as: Le, K. D., Nguyen, H. D., Ranmuthugala, D., Forrest, A. L., 2016. Artificial potential field for ROV haptic control in dynamic environment, Proceedings of the Institution of Mechanical Engineers, Part I: Journal of systems and control engineering, 230(9), 962-977