FxT_VFO

A group of cooperating vehicles can often perform tasks that are difficult or even infeasible for an individual agent, while increasing robustness to potential vehicle failures and flexibility. Therefore, the design of navigation schemes for autonomous multi-vehicle systems is of practical interest (e.g., surveillance, transportation, ...). Despite an abundant literature on the subject, several theoretical and technical challenges remain to be solved (e.g. input saturation, time constraints, non-holonomy, curvature limits, collision avoidance, ...). It therefore seems crucial to design a new distributed navigation system for multi-robot systems taking into account the previously mentioned constraints.

It may be mentioned that navigation of non-holonomic systems raises significant challenges due to non-integrable constraints. Indeed, this class of systems cannot be stabilized by a continuous state feedback controller. To solve this problem, a vector field orientation (VFO) control methodology has been introduced without considering temporal constraints.

This thesis aims to design new navigation schemes for non-holonomic mobile robots using the vector field orientation (VFO) approach. The main impact is expected in the removal of some locks concerning the consideration of constraints (input saturation, time constraints, curvature constraints, collision avoidance, ...) for non-holonomic systems.

This thesis aims at designing new navigation schemes for mobile non-holonomic robots using the vector field orientation (VFO) approach.