Transactions of the Institute of Measurement and Control

Hong Wang, ; Nokbakhti, A.; Cook, Peter; Owens, D.H.

doi: 10.1177/0142331207083469pmid: N/A

Zhengtao Ding, ; Zairong Xi, ; Hong Wang,

doi: 10.1177/0142331207083485pmid: N/A

This paper provides a brief review of Rosenbrock's recent work on generating quantum mechanics using dynamic programming, some background knowledge of quantum mechanics and some general issues and problems in developing feedback quantum control.

Rosenbrock, H.H.; Ding, Z.

doi: 10.1177/0142331207083484pmid: N/A

This brief paper continues a development in earlier publications by Rosenbrock, in which results in quantum mechanics were obtained by adding a random disturbance to Hamilton's principle. The disturbance is complex, and the resulting variational problem is solved by dynamic programming, which computes the causal relations needed to accomplish a purpose. It is shown here that this procedure leads to many results that agree with those obtained in the standard theory of quantum mechanics, but also to some that are not readily obtained in that theory. For example, certain changes at a time tm to the end condition xf, which will be reached at tf>tm, can instantaneously change some aspects of the particle's behaviour at t m while it is still in transit. This accounts for Einstein's `spooky action at a distance'. An approach is also given to interference and entanglement, which is simpler than the standard account and may therefore be useful in quantum control and quantum computation.

Takahashi, Yoshihiko; Ogawa, Shinobu; Machida, Shigenori

doi: 10.1177/0142331206075538pmid: N/A

A robotic wheelchair with inverse pendulum control has been proposed in this paper. By using the robotic wheelchair, a wheelchair-bound person can climb over a step up to about 7 cm without a helper's assistance. The robotic wheelchair moves on flat ground using four wheels. It raises its front wheels in front of a step, and maintains the inverse pendulum control. A user riding can go forwards and backwards using his/her body inclination maintaining the inverse pendulum control. The inverse pendulum control is stopped when the front wheels proceed over a step. After the front wheels land on the step, the robotic wheelchair can climb over the step, assisted by DC motor force. In order to raise the front wheels using small DC motors, the robotic wheelchair can move its rear wheel drive shafts. This paper presents the experiments on step climbing, the modelling of the control system, and the simulations on the inverse pendulum control. The step climbing function of the proposed robotic wheelchair has been successfully tested.

Boršč, M.; Vitko, A.; Thurský, B.

doi: 10.1177/0142331207075586pmid: N/A

The paper shows that a control strategy based on the changing structure of a control system, ie, on controlled jump changes of the control system parameters, ensures more rapid stabilization than a control system with constant parameters. An algorithm of the time-optimal stabilization is designed in accordance with the Pontriagin's maximum principle in such a way that the resulting behaviour is non-periodic. The designed control algorithm divides the state space into segments in which the parameters take their limit values. The control problem is resolved for third-order and higher linear systems. Issues of system robustness are demonstrated by the example of stabilizing a variably loaded gantry crane.

Zhang, L.; Bai, Y.F.

doi: 10.1177/0142331207076374pmid: N/A

The paper presents a new method to achieve maximum power point tracking (MPPT) for practical grid-connected PV panels. The method employs the radial basis function neural network (RBFNN) to predict the PV plant's maximum power points corresponding to different weather conditions. The RBFNN model can be trained on-line, autonomously, using a simplified genetic algorithm (SGA). The method has been verified by modelling the MPP points for practical PV panels located in Southampton and Leeds, respectively. The Leeds model has been used to control a prototype PV-grid-connected power generation system. The on-line RBFNN training scheme is discussed in detail and experimental results are presented which compared favourably with the conventional perturbation and observation (P&O) technique.