Clustering methods assign objects to clusters using only as prior information the characteristics of the objects. However, clustering algorithms performance can be improved when background knowledge is available. Such background knowledge can be incorporated in a clustering method as label constraints which results in a semi-supervised clustering algorithm. We propose to extend two possibilistic clustering algorithms to make use of available a priori information. The goal is twofold: to improve the accuracy of the clustering result by leading the method towards a desired solution and to detect outliers by taking advantage of the generated possibilistic partition. The proposed methods are called semi-supervised repulsive possibilistic c-means (SRPCM) and semi-supervised possibilistic fuzzy c-means (SPFCM). They correspond to possibilistic clustering algorithms that introduce label constraints. Experimental results show that the proposed algorithms using label constraints improve (1) the clustering result and (2) the outliers detection.

This paper addresses the problem of forced bipartite consensus on multi-agent systems (MAS) with application to quadrotor coordination. Opposite to standard consensus algorithms, the purpose of bipartite consensus is to divide the MAS into two groups or clusters. Resulting groups or clusters converge to the exact same solution with opposite sign, i.e. agents in the same group find a consensus or agreement while clusters, between them, get to a disagreed consensus. We extend the concept of multi-agent bipartite consensus to the case of forced bipartite consensus on leader-based multi-agent systems. Since tracking of a desired reference is based on the center of mass of the entire group, we analyze the controllability and observability properties from input and output of the leader. Quadrotor platoon switching behavior between coordination and danger avoidance shows a potential application of the proposed control approach. Results have been validated in simulation.

In this paper, the use of a class of robust control technique as a tool to compute the robustness margin for a closed loop multi-agent system is discussed. Our approach is based on a leader/follower structure of the multiple robot systems. It is shown that, for multiple agent system with cyclic, chain and balanced topology of information exchange, the result is achieved. A Lyapunov based robust control design for coordination and trajectory tracking of multiple agent systems with parametric uncertainty is presented. The results are illustrated with examples and simulations.

This paper addresses the problem of time optimal path planning for a quadrotor helicopter evolving in a region of known winds. Usually, the flight control of quadrotors subject to wind disturbances challenge seeks to find the optimal control to keep track of a desired trajectory in a windy region. This approach has one major disadvantage: the quadrotor flight control has to compensate for trajectory deviations; therefore, the energy consumption becomes an issue. Most unmanned aerial vehicles (UAV) navigation techniques use waypoints to accomplish their missions. In the framework of a waypoint based navigation, a promising path planning strategy would be a time optimal approach in which the UAV would take advantage of wind to reach its next waypoint; therefore, saving time and energy (under constant forward velocity constraint). A model separation is used to simplify the control of the six-degrees-of-freedom (6DOF) dynamics of the quadrotor. Such approach allows to deal with quadrotor's 3D-motion through two subsystems: dynamic (altitude and MAV-relative forward velocity) and kinematic (nonholonomic navigation) subsystems. In terms of control, a hierarchical control scheme is used to stabilize dynamic and kinematic underactuated subsystems involved in the navigation task. The time optimal path planning is computed using a dynamic optimization method for continuous systems with some state variables specified at an unspecified terminal time. Results have been validated in simulation.

In this paper, the coordination and trajectory tracking control design of multiple mini rotorcraft systems are discussed. The dynamic model of a mini rotorcraft is presented using the Newton-Euler formalism. Our approach is based on a leader/follower structure of multiple robot systems. The centroid of the coordinated control subsystem is used for trajectory tracking purposes. A nonlinear coordinated control design for multiple autonomous vehicle synchronization is developed. The analytic results are supported by simulation tests.

This paper addresses the coordination and trajectory tracking control design for a multiple mini rotorcraft system. The dynamic model of the mini rotorcraft is presented using the Newton-Euler formalism. Our approach is based on a leader/follower structure of multiple robot systems. A nonlinear controller based on nested saturations and a multi-agent consensus control are combined to obtain a flight formation control for a multiple mini rotorcraft system. The centroid of the coordinated control subsystem is used for trajectory tracking purposes. The analytic results are supported by simulation tests.

In this paper, the modeling and control design of a vertical flight attitude control for a mini tail-sitter with variable pitch propeller is discussed. Tail-sitters VTOL-UAVs have operational flexibility of typical helicopters while having the cruise performance of fixed wing airplanes. The configuration proposed in this work is highly unstable in its natural flight state in vertical mode. A dynamic model is developed and a linear control strategy has been developed to stabilize the platform. The results are supported by experimental tests.