AutoNOMO mobile robot control using variable structure approaches

Autonomous (self-driving) cars are complex systems capable of sensing the environment and making decisions to move without human intervention. Before autonomous cars become entirely functional, intensive research work and tests are necessary. In this context, the use of scaling vehicles for testing and verification is customary [1]. Scaled vehicles are a useful tool to find drawbacks in the interconnected sensing and control loops before their deployment into a full-scale vehicle.

The AutoNOMO vehicle is devoted to research and educational purposes. The AutoNOMO is a car-like robot comparable to a full-scale autonomous vehicle in kinematic, sensing, and computation skills. It is furnished with all the sensors available in a full-scale vehicle, i.e., rotating laser scanner RPLIDAR, accelerometers, gyroscopes, encoders, camera, and an indoor GPS (i.e., fish-eye camera), see [2] for further details.

In terms of navigation and control, AutoNOMO can be considered a car-like robot. Car-like robots are nonlinear underactuated systems with nonholonomic restrictions. The control of such systems is a challenging problem. Such a nonholonomic system may be represented in the so-called chained-form, which possesses homogeneity properties. Therefore, the design of homogeneous controllers that preserve the system homogeneity becomes desirable. Moreover, homogeneous systems of negative degree present finite-time convergence properties that improve the system navigation.

Homogeneity is a mathematical property that rends the system more treatable in terms of control design. Therefore, the thesis aims to design homogeneous controllers that solve, in a simple and elegant way, and with low computational burden, the trajectory tracking problem of a car-like system.

AutoNOMO

General work planning:

- Revision of the state of art

- Analysis of system structure (controllability, feasibility of trajectories, etc.)

- Stability analysis

- Experimental validation

- Writing of scientific articles

Framework:

The Master thesis will be co-supervised in the frame of the international project ‘’Robust control paradigm for nonholonomic systems: a variable structure approach’’.

Candidate profile:

- B. Eng. Or B. Sc. Degree in systems and control, electrical, electronic, mechatronic or aerospace, etc.

- Good English level.

- Strong analytical and communications skills.

How to apply

Candidates must send a detailed CV and motivation letter to dferreira@citedi.mx. The full instructions for the postgraduate call can be checked at https://maestria.citedi.mx

Starting date:

August or February

Grant:

Accepted students may apply for CONACyT financial support, and IPN travelling support.

Workplace

Tijuana, B.C., Mx.

Contact:

Dra. Alejandra Ferreira de Loza (dferreira@citedi.mx) https://www.researchgate.net/profile/Alejandra_Ferreira_de_Loza

References:

1) A. Berntorp, and T. Hoang, and R. Quirynen, and S. Di Cairano, “Control Architecture design for autonomous vehicles”, 2018 IEEE Conference on Control Technology and Applications (CCTA), 2018, pp. 404–411.

2) The AutoNOMOS Model website. [Online]. Available: https://github.com/AutoModelCar/AutoModelCarWiki/wiki/Hardware.

3) A. de Luca, and G. Oriolo, and C. Samson, Feedback Control of a Nonholonomic Car-like Robot. In J.P. Laumond (ed.), Robot motion planning and control, Springer, 1998.

4) S. Kimura, and H. Nakamura, and Y. Yamashita, “Asymptotic stabi[1]lization of nonholonomic four-wheeled vehicle with steering limitation based on Lyapunov function approach,” IFAC Nonlinear Control Systems Symposium, 2016, pp.235–240.

5) A. Rosas-Vilchis, A. F. de Loza, L. T. Aguilar, J. Cieslak, D. Henry and O. Montiel-Ross, "Trajectory tracking control for an AutoNOMO vehicle using a decoupling approach," 2020 28th Mediterranean Conference on Control and Automation (MED), 2020, pp. 375-380, doi: 10.1109/MED48518.2020.9183255.