Trajectory Planning of the Humanoid Manipulator

The paper presents a method of planning a collision-free trajectory for a humanoid manipulator mounted on a rail system. The task of the robot is to move its end-effectors from the current position to the given final location in the workspace. The method is based on a redundancy resolution at the velocity level. In addition to this primary task, secondary objectives are also taken into account. The motion of the robot is planned in order to maximize a manipulability measure in purpose of avoiding manipulator singularities. State inequality constraints resulting from collision avoidance conditions are also considered. A computer example involving a humanoid manipulator operating in a three dimensional task space is also presented.

Optimal impulsive control for advertising strategy problems based on a gradient-based PSO algorithm

This paper considers an optimal advertising strategy problem. This is an important problem in marketing investment for new products in a free market. The main contributions of this paper are as follows. First, the problem is formulated as an optimal control problem of switched impulsive systems with the state inequality constraints, which is different from the existing nonlinear system models. As the complexity of such constraints and the switching instants are unknown, it is difficult to solve this problem by using conventional optimization techniques. To overcome this difficulty, by applying the penalty function, all the state inequality constraints are first written as non-differentiable penalty terms and imposed into the cost function. Then, the penalty terms are smoothed by using a novel smooth function, leading to a smooth optimal control problem with no state inequality constraints, and an improved gradient-based particle swarm optimization (PSO) algorithm is proposed for solving this problem. Error analysis results show that if the adjustable parameter is sufficiently small, the solution of the smooth optimal control problem is approximately equal to the original problem. Finally, a switched impulsive system for beer sales is established to illustrate the effectiveness of the developed algorithm.