This thesis addresses the software development and systems upgrade of a colaborative mobile robot cell, developed at the Laboratory of robotics. The mobile cell is composed of a robotic mobile platform, a colaborative robot manipulator and various supporting systems, that handle power, safety and communications. The thesis describes the design and implementation of power delivery systems, integration of sensor systems and development of high- and low-level software. As part of the thesis, a simple simulation of the platform was subsequently developed, due to ongoing emergency measures during the COVID-19 outbreak, which prevented the testing of the robot cell. The final goal of the thesis is to implement the necessary control algorithms, realize autonomous navigation and prepare the system for further development.
The thesis first focuses on the theoretical background, required for the proposed upgrade. In terms of hardware, it describes the kinematic model of the drive system and platform. The second half focuses more on software, and describes the working principles of popular navigational and SLAM algorithms.
The thesis continues with the description of the hardware upgrades. We equipped the cell with a battery system, for use in autonomous navigation and a power supply, which is to be used during development when the platform is connected to the power grid. We also added a high performance computer capable of virtualisation and laser scanners, witch will be used for safety and autonomous navigation. Lastly, we upgraded the existing communication network.
The next part describes the development and structure of newly implemented software. The low-level software was developed on an industrial controller, while the high-level software runs on the newly installed high performance computer with the help of virtualisation. They communicate with simple binary UDP packets. The low-level software includes algorithms for calculating platform kinematics and odometry, a communication interface, a manual control joystick interface and safety features. In terms of high-level software, we mainly focused on integrating the virtualization environment and implementing autonomous driving with the help of ROS.
In the last part, the thesis covers the platform simulation, which was used for development and testing of high-level software. The simulation was made with the help of Unity and includes an emulation of the communication interface, implemented on the industrial controller. We can therefore test all high-level software without any modifications. With it, we tested multiple ROS packages for mapping and navigation that are available on the ROS repository and made a short summary, describing their main features. We also discussed their advantages and disadvantages in regards to our application.
In the conclusion we review and evaluate the goals of the thesis. We also mention the problems that made development difficult. We finish by proposing potential improvements and presenting the direction of further development.