The master's thesis presents the development, simulation, and implementation of a control system for a robotic vehicle capable of performing an aerial backflip during a jump and achieving a stable landing. The work includes the identification of the real system and the subsequent development of a corresponding simulation model, the planning of the aerial maneuver, the design of the control algorithm, and the transfer of the developed solution from the simulation environment to the real system.
In the simulation part, a vehicle model was developed in the Gazebo environment, enabling the design and evaluation of different wheel speed profiles during the aerial maneuver. By incorporating a PD controller, the stability of the rotation and the accuracy of the landing were improved. The system demonstrated a high level of repeatability and robustness, even in the presence of measurement noise and minor external disturbances.
In the practical part, the system was transferred to a real platform based on the Arduino Nano microcontroller and a Raspberry Pi computer. An interface was developed for experiment control, status detection, and safety mechanisms to prevent damage to the device. Experimental results showed that the vehicle successfully performs multiple consecutive jumps with rotation and landing, confirming the correctness of the controller design and the appropriateness of the proposed approach.
|
