Design and implementation of a Sensory System for an Autonomous Mobile Robot in a Connected Industrial Environment

Technological evolution has brought considerable progress in the field of automation by introducing and developing systems that have become essential for operations in the industrial field. The development of autonomous systems has allowed the facilitation of human work by allowing collaboration with it but also a replacement to carry out the heaviest tasks. The purpose of this thesis is to develop a sensory system for an Autonomous Mobile Robot (AMR) capable of moving in an industrial environment sharing space with human operators. To do this, a complex and robust system has been developed consisting of a set of sensors suitable for to the environment in which the robot should move. Exploiting the perceptions of the onboard sensors it has been possible to obtain an autonomous navigation system. For the choice and connection of the sensors, the state of the art of the most used navigation systems has been studied taking into account a set of fundamental aspects, as well as some problems that could have limited the navigation system. It consists of the Scout Mini platform made up of four Mecanum wheels enriched by a larger structure used for maintenance purposes in an industrial environment. The latter limits the range of vision and therefore the omnidirectional movement allowed by the type of wheels used. In addition, the choice of sensors also takes into account the errors introduced by these types of wheels that are subject to vibration and slippage. The system, developed using ROS (Robot Operating System), has been simulated and experimentally implemented by choosing an appropriate position of the sensors chosen, an architecture that allows the reduction of computational cost and using some algorithms to fuse and filter data from sensors.