C. Frangos

Computation of feasible control trajectories for the navigation of a ship around an obstacle in the presence of a sea current

A ship has to move from a point A to a point B and, on its way, has to circumvent an obstacle. In addition, the presence of a sea current is assumed. Using a realistic model of a tanker ship, a method is proposed for computing feasible control trajectories for the navigation of the ship.

Penalty function solutions to optimal control problems with general constraints via a dynamic optimisation method

Abstract The optimal control problem with general constraints is considered. This involves the calculation of the open loop control input as a function of time such that an objective function is minimised and the constraints are satisfied. The approach taken in this paper is to discretise the control input by representing it by a sequence of piecewise constant vectors over equally spaced intervals spanning a finite time horizon. A penalty function approach is then used to reformulate the original constrained problem as an unconstrained minimisation problem which is solved by using a dynamic minimisation algorithm. This algorithm is particularly suitable for penalty functions. The approximate solution obtained minimises the objective function whilst satisfying the original constraints at the discretisation points.

Feasible command strategies for the control of a rolling disk with a rod fixed along its axis

The concept of feasible command strategies is introduced and their applicability is demonstrated by solving a guidance and control problem. This problem concerns the motion of a system which is composed of a rolling disk and a slender rod that is fixed through the disks center along the disks axis. The motion of the disk-rod system is subjected to state and control constraints, and it can serve as an extension to a simplified model of the motion of a riderless unicycle.

The performance of a projectile which uses a bang-bang type guidance law: Part 1

Abstract A projectile which uses a bang-bang type guidance law is launched, and its goal is to hit a fixed target whose centre is located on the ground. Using stochastic optimal control, the performance of the projectile is compared with cases where an optimal guidance law, or a saturated proportional navigation law, are being applied using the same airframe.

The motion of a cylinder-rod system on a horizontal plane: Feasible control and path controllability

This work deals with the guidance and control of a system which is composed of a rolling cylinder, and a controlled slender rod that is pivoted through its center of mass, about the cylinders center. At some instant the rod has to establish contact with a smooth rigid ceiling. This instant constitutes a discontinuity in the systems dynamics. The cylinder-rod system is subjected to state and control constraints, and it serves as a model for the motion of a simple mobile robot. Open-loop strategies are computed for the cylinder-rod system. In addition, the concept of path controllability is introduced for this system, and a condition is derived for the systems motion path controllability. The derivation of this condition enables one to design closed-loop control laws for the systems motion.

The performance of a projectile which uses a bang-bang type guidance law: Part 2—A semirigid body model

Abstract A projectile which uses a bang-bang type guidance law is launched and its goal is to hit a fixed target whose center is located on the ground. In [1], applying stochastic control methods the performance of the projectile is investigated by using a point-mass model. In this work, which is a sequel to [1], a semirigid body model is used for the projectile and its performance is investigated by using stochastic control methods.

Optimal selection and control strategies for a flexible manufacturing and assembly system

Abstract This paper deals with the selection-control problem of a system which consists of a flexible manufacturing cell, M production lines and an assembly machine. The flexible manufacturing cell feeds the production lines, and the production lines feed the assembly machine. Sufficient conditions on optimal feedback selection-control strategies, that maximize the expected value of the total time that all the buffers concerned are simultaneously nonempty, are derived. Also, a simple feedback selection-control strategy is proposed.

An optimal selection policy for a flexible manufacturing cell feeding several production lines

Abstract This paper deals with the selection problem of a flexible manufacturing cell feeding several production lines. Sufficient conditions on optimal selection policies, which maximize the total amount of time that all the production lines are simultaneously active, are derived. A procedure for computing a suboptimal feedback selection policy is also proposed.

Control strategies for a stochastic flexible manufacturing and assembly system model

Abstract Optimal control strategies are computed for a stochastic discrete time flexible manufacturing and assembly system model. An easy to implement feedback control strategy is also proposed with a cost that compares favourably with the optimal cost.

The performance of some manufacturing queueing networks when some of the machines are in the failure mode

Abstract Using the theory of right-continuous Markov processes with values in N+k, an optimal control problem concerning the performance of manufacturing queueing networks, when some of the machines are in the failure mode, is dealt with.