Particle swarm optimization based trajectory tracking control of a proton exchange membrane fuel cell driven automated guided vehicle

Automatic guided vehicles (AGV) are driverless, mobile vehicles programmed to transport materials in a facility. They are used in many warehouse and material transfer applications in many industries to compensate for employee deficiencies, reduce labor costs and increase warehouse efficiency. AGVs significantly reduce the damage of goods and industrial accidents during the process by minimizing human interactions with the transported loads. AGVs have power systems that store energy to make mobile work possible and convert it into mechanical rotation or displacecement energy at the desired rate. These power systems include batteries that store energy and motors that take energy from the battery and convert it into mechanical energy Regardless of their size mobile robots must carry enough energy to achive useful uptime in the field before re-charging the battery. This concept, known as energetic autonomy, includes two criteria: the ability to operate for a useful period of time and the ability to oparate independently of an external power source. For this purpose, for the study of AGVs, a type of mobile robot, this study focused on a alternative energy-derived drive syytem, and the design of a fuel cell-driven AGV and analysis of this AGV were performed. Fuel cells use hydrogen as a fuel and provide the advantage of a longer working time for AGVs as an alternative energy source that generates energy. In this study, firstly a moduler AGV design that will meet the desired requirements and create an environment suitable for carrying loads has been realized. Structural analyzes were made to determine the load capacity the the system, which has mecanum wheels and DC motors to drive these wheels, and the final design was achieved by improving the strength of the system. In addititon, the kinematic model was created in the global coordinate system of AGV and the coordinate system create for each wheel and the kinematic model of the system, dynamic model and Langrange energy equations using linear velocity and angular velocity changes for the AGV and for each wheel. Along with these models, the PEM cuel cell model, which will be used as a battery system, was also created and real-time tracking of the system's specific trajetory was performed using the overall system model obtained.