Kingsley Fregene

Decentralized Receding Horizon Control and Coordination of Autonomous Vehicle Formations

This paper describes the application of a novel methodology for high-level control and coordination of autonomous vehicle teams and its demonstration on high-fidelity models of the organic air vehicle developed at Honeywell Laboratories. The scheme employs decentralized receding horizon controllers that reside on each vehicle to achieve coordination among team members. An appropriate graph structure describes the underlying communication topology between the vehicles. On each vehicle, information about neighbors is used to predict their behavior and plan conflict-free trajectories that maintain coordination and achieve team objectives. When feasibility of the decentralized control is lost, collision avoidance is ensured by invoking emergency maneuvers that are computed via invariant set theory.

Distributed Cooperative Outdoor Multirobot Localization and Mapping

The subject of this article is a scheme for distributed outdoor localization of a team of robots and the use of the robot team for outdoor terrain mapping. Localization is accomplished via Extended Kalman Filtering (EKF). In the distributed EKF-based scheme for localization, heterogeneity of the available sensors is exploited in the absence or degradation of absolute sensors aboard the team members. The terrain mapping technique then utilizes localization information to facilitate the fusion of vision-based range information of environmental features with changes in elevation profile across the terrain. The result is a terrain matrix from which a metric map is then generated. The proposed algorithms are implemented using field data obtained from a team of robots traversing an uneven outdoor terrain.

Distributed heterogeneous outdoor multi-robot localization

An extended Kalman filter (EKF)-based algorithm for the localization of a team of robots is described in this paper. The distributed EKF localization scheme is straightforward in that the individual robots maintain a pose estimate using EKFs that are local to every robot. We then show how these results can be extended to perform heterogeneous cooperative localization in the absence or degradation of absolute sensors aboard the team members. The proposed algorithms are implemented using field data obtained from a team of ATRV-mini robots traversing on uneven outdoor terrain.

Hybrid decentralized control of large scale systems

Motivated by three applications which are under investigation at the Honeywell Research Laboratory in Minneapolis, we introduce a class of large scale control problems. In particular we show that a formation flight problem, a paper machine control problem and the coordination of cameras in a monitoring network can be cast into this class. In the second part of the paper we propose a decentralized control scheme to tackle the complexity of the problem. The scheme makes use of logic rules which improve stability and feasibility of the decentralized method by enforcing coordination. The decentralized control laws which respect the rules are computed using hybrid control design.

Stabilizing control of a high-order generator model by adaptive feedback linearization

We present an adaptive feedback linearizing control scheme for excitation control and power system stabilization. The power system is a synchronous generator which is first modeled as an input-output nonlinear discrete-time system approximated by two neural networks. Then, the controller is synthesized to adaptively compute an appropriate feedback linearizing control law at each sampling instant using estimates provided by the neural system model. This formulation simplifies the problem to that of designing a linear pole-placement controller which is itself not a neural network but is adaptive in the sense that the neural estimator adapts itself online. Additionally, the requirement for exact knowledge of the system dynamics, full state measurement, as well as other difficulties associated with feedback linearizing control for power systems are avoided in this approach. Simulations demonstrate its application to a high-order single-machine system under various conditions.

Toward a systems- and control-oriented agent framework

This paper develops a systems- and control-oriented intelligent agent framework called the hybrid intelligent control agent (HICA), as well as its composition into specific kinds of multiagent systems. HICA is essentially developed around a hybrid control system core so that knowledge-based planning and coordination can be integrated with verified hybrid control primitives to achieve the coordinated control of multiple multimode dynamical systems. The scheme is applied to the control of teams of unmanned air and ground vehicles engaged in a pursuit-evasion war game. Results are demonstrated in simulation.

Decentralized Receding Horizon Control of Cooperative Vechicle Formations

An overview of a novel approach to vehicle formation control is presented. The scheme is based on decentralized and cooperative predictive control. The control scheme is being tested on Unmanned Air Vehicle models at the Honeywell Laboratories in Minneapolis. Each vehicle is equipped with efficient guidance and control loops which enable the implementation of the decentralized scheme for higher-level control and coordination. In this framework, collision avoidance and constraint fulfillment are ensured by using a logic that switches to collision-free emergency maneuvers if necessary. The size of protection zones associated with this logic is determined from invariant set theory. The scheme also makes use of logic rules which improve stability and feasibility of the decentralized method by enforcing coordination. The decentralized control laws which respect the rules are computed using hybrid control design techniques. The proposed decentralized control scheme is formulated as optimization problems of small sizes which can be translated into equivalent piecewise affine state-feedback controllers. These controllers can then be downloaded as corresponding look-up tables to the onboard avionics of the vehicle and run in realtime.

Distributed Heterogeneous Sensing for Outdoor Multi-Robot Localization, Mapping, and Path Planning

Our objective is to develop a team of autonomous mobile robots that are able to operate in previously unfamiliar outdoor environments. In these environments, the robot teams should be able to cooperatively localize even when DGPS is not consistently available, to autonomously generate rough elevation maps of their terrain, and to use these generated maps to plan multi-robot paths that enable them to accomplish their mission objective, such as reconnaissance and surveillance or perimeter security. This paper briefly outlines our approaches to achieving this objective, along with some of our implementation results on our team of four ATRV-mini mobile robots.

Incremental multi-agent robotic mapping of outdoor terrains

We describe a scheme for building terrain maps of realistic outdoor environments by having multiple robotic agents navigate in them. The terrain map combines vision-based depth information of environmental features with an elevation gradient created by fusing vertical displacements obtained from inclinometer pitch angles with DGPS altitude data. Experimental results are presented to illustrate the practical application of this scheme.

Autonomous tracking of a ground vehicle by a UAV

We propose a generally applicable method for the tracking of ground vehicles by aerial vehicles. The method does not require any modifications to the guidance and control of the UAV. It only requires the capability to follow waypoint commands. Sensing of ground vehicle position with significant time delays is assumed. The delays model the time of image processing, and the communication delays involved in sending data to a ground station, performing the computations and receiving the results on the UAV. Variants of the method have been tested successfully on the field and may see widespread deployment.