Date of Award


Publication Type


Degree Name



Electrical and Computer Engineering


Control and Estimation;Robotics;Special Euclidean Group;UGV and UAV


Xiang Chen



Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


This dissertation presents a controlling method that utilizes visual observation to address the challenge of driving an autonomous vehicle with a kinematic model on SE(3), by using the data obtained from the onboard cameras. This controlling framework is applicable when direct retrieval of the vehicle's attitude information is not possible or when precision control of the vehicle is necessary. However, addressing several challenges to attain closed-loop stability for the observer-based control for autonomous vehicles on SE(3) is essential. 1. The analysis is further complicated by the presence of multiple equilibria in the error dynamics associated with the suggested method. 2. The presence of inherent uncertainty in the kinematic model of the vehicle is an inevitable factor. 3. The existence of geometric constraints inside the camera system results in output switching and the possibility of misidentification. 4. The presence of measurement noise from the camera introduces supplementary uncertainty into the system. 5. The presence of nonlinearity within the system presents significant challenges when it comes to the development of an observer and controller. This dissertation employs a distinct design approach for the observers and controllers, and ensures closed-loop stability by utilizing the well-known small gain theorem. When developing the observer, the geometrical constraint of the sensor is parameterized and addressed in observer design. In addition, multiple equilibria of the error dynamics can be resolved by employing a carefully chosen gain, achieving semi-global practical asymptotic stability. Furthermore, for the observer, the semi-global practical asymptotic stability can be relaxed to semi-global practical input-to-state stability in the presence of measurement noise and modeling uncertainty. The controller is also developed, assuming the system state can be retrieved. The achievement of semi-global practical asymptotic stability is also guaranteed by carefully selecting the gain to mitigate the possibility of multiple equilibria. The achievement of semi-global practical input-to-state stability for the controller can be guaranteed by increasing the controller's gain in the presence of the modeling uncertainty and measurement noise. With the two guaranteed semi-global practical input-to-state stable systems, the observer and the controller, the semi-global practical input-to-state stability can be achieved for the closed-loop systems, while the linear separation principle does not apply to the two nonlinear subsystems. The simulation and experiments are conducted to verify the suggested design of the observers and controllers and assess the stability of the closed-loop system. Furthermore, a thorough analysis of the existing literature, when compared to the proposed research, demonstrates the unique advantages of the proposed approach in this study. In addition, this paper presents a set of tuning instructions for optimizing the performance of the suggested observer.