Jianqiang Yi

A computed torque controller for uncertain robotic manipulator systems: Fuzzy approach

Computed Torque Control (CTC) is an effective motion control strategy for robotic manipulator systems, which can ensure globally asymptotic stability. However, CTC scheme requires precise dynamical models of robotic manipulators. To handle this impossibility, in this paper, a new approach combing CTC and Fuzzy Control (FC) is developed for trajectory tracking problems of robotic manipulators with structured uncertainty and/or unstructured uncertainty. Fuzzy part with a set of tunable parameters is employed to approximate lumped uncertainty due to parameters variations, unmodeled dynamics and so on in robotic manipulators. Based on Lyapunov stability theorem, it is shown that the proposed controller can guarantee stability of closed-loop systems and satisfactory tracking performances. The proposed approach indicates that CTC method is also valid for controlling uncertain robotic manipulators as long as compensative controller is appropriately designed. Finally, computer simulation results on a two-link elbow planar robotic manipulator are presented to show tracking capability and effectiveness of the proposed scheme.

Anti-swing and positioning control of overhead traveling crane

A new fuzzy controller for anti-swing and position control of an overhead traveling crane is proposed based on the Single Input Rule Modules (SIRMs) dynamically connected fuzzy inference model. The trolley position and velocity, the rope swing angle and angular velocity are selected as the input items, and the trolley acceleration as the output item. Each input item is given with a SIRM and a dynamic importance degree. The control system is proved to be asymptotically stable to the destination. The controller is robust to different rope lengths and has generalization ability for different initial positions. Control simulation results show that by using the fuzzy controller, the crane is smoothly driven to the destination in short time with small swing angle and almost no overshoot.

Trajectory Tracking Control of Omnidirectional Wheeled Mobile Manipulators: Robust Neural Network-Based Sliding Mode Approach

This paper addresses the robust trajectory tracking problem for a redundantly actuated omnidirectional mobile manipulator in the presence of uncertainties and disturbances. The development of control algorithms is based on sliding mode control (SMC) technique. First, a dynamic model is derived based on the practical omnidirectional mobile manipulator system. Then, a SMC scheme, based on the fixed large upper boundedness of the system dynamics (FLUBSMC), is designed to ensure trajectory tracking of the closed-loop system. However, the FLUBSMC scheme has inherent deficiency, which needs computing the upper boundedness of the system dynamics, and may cause high noise amplification and high control cost, particularly for the complex dynamics of the omnidirectional mobile manipulator system. Therefore, a robust neural network (NN)-based sliding mode controller (NNSMC), which uses an NN to identify the unstructured system dynamics directly, is further proposed to overcome the disadvantages of FLUBSMC and reduce the online computing burden of conventional NN adaptive controllers. Using learning ability of NN, NNSMC can coordinately control the omnidirectional mobile platform and the mounted manipulator with different dynamics effectively. The stability of the closed-loop system, the convergence of the NN weight-updating process, and the boundedness of the NN weight estimation errors are all strictly guaranteed. Then, in order to accelerate the NN learning efficiency, a partitioned NN structure is applied. Finally, simulation examples are given to demonstrate the proposed NNSMC approach can guarantee the whole systems convergence to the desired manifold with prescribed performance.

Neural Network Control for a Semi-Active Vehicle Suspension with a Magnetorheological Damper:

Semi-active vehicle suspension with magnetorheological dampers is a promising technology for improving the ride comfort of a ground vehicle. However, the magnetorheological damper always exhibits nonlinear hysteresis between its output force and relative velocity, and additional nonlinear stiffness owing to the state transition from liquid to semi-solid or solid, so that the semi-active suspension with magnetorheological dampers features nonlinearity by nature. To control such nonlinear dynamic systems subject to random road roughness, in this paper we present a neural network control, which includes an error back propagation algorithm with quadratic momentum of the multilayer forward neural networks. Both the low frequency of road-induced vibration of the vehicle body and the fast response of the magnetorheological damper enable the neural network control to work effectively on-line. The numerical simulations and an experiment for a quarter-car model indicate that the semi-active suspension with a magnetorheological damper and neural network control is superior to the passive suspensions in a range of low frequency.

Stabilization fuzzy control of inverted pendulum systems

A new fuzzy controller for stabilization control of inverted pendulum systems is presented based on the Single Input Rule Modules (SIRMs) dynamically connected fuzzy inference model. The fuzzy controller has four input items, each with a SIRM and a dynamic importance degree. The SIRMs and the dynamic importance degrees are designed such that pendulum angular control has priority over cart position control. It is made clear that the fuzzy controller performs the pendulum angular control and the cart position control in parallel, and switching between the two controls is realized by automatically tuning the dynamic importance degrees according to control situations. The simulation results show that the proposed fuzzy controller has a high generalization ability to stabilize completely a wide range of the inverted pendulum systems within 9.0 s for an initial angle up to 30.08. q 2000 Elsevier Science Ltd. All rights reserved.

A new fuzzy controller for stabilization of parallel-type double inverted pendulum system

A new fuzzy controller with 6 input items and 1 output item for stabilizing a parallel-type double inverted pendulum system is presented based on the single input rule modules (SIRMs) dynamically connected fuzzy inference model. Each input item is assigned with a SIRM and a dynamic importance degree. The SIRMs and the dynamic importance degrees are designed such that the angular control of the longer pendulum takes the highest priority over the angular control of the shorter pendulum and the position control of the cart when the angle of the longer pendulum is big. By using the SIRMs and the dynamic importance degrees, the priority orders of the three controls are automatically adjusted according to control situations. The proposed fuzzy controller has a simple and intuitively understandable structure, and executes the three controls entirely in parallel. Simulation results show that the proposed fuzzy controller can stabilize completely a parallel-type double inverted pendulum system within 10.0 s for a wide range of the initial angles of the two pendulums. This is the first result for a fuzzy controller to achieve successfully complete stabilization control of a parallel-type double inverted pendulum system.

A proposal of SIRMs dynamically connected fuzzy inference model for plural input fuzzy control

Single input rule modules (SIRMs) dynamically connected fuzzy inference model is proposed for plural input fuzzy control. For each input item, a SIRM is constructed and a dynamic importance degree is defined. The dynamic importance degree consists of a base value insuring the role of the input item through a control process, and a dynamic value changing with control situations to adjust the dynamic importance degree. Each dynamic value can be easily tuned based on the local information of current state. The model output is obtained by summarizing the products of the dynamic importance degree and the fuzzy inference result of each SIRM. The controller constructing method for constant value control systems is given, and constant value controls of typical first- and second-order lag plants are tested. The simulation results show that by using the proposed mode, the reaching time can be reduced by more than 15% without any steady-state error, overshoot, or vibration compared with the SIRMs fixed importance degree connected fuzzy inference model. The proposed model is further successfully applied to stabilization control of an inverted pendulum system including the position control of the cart.

Upswing and stabilization control of inverted pendulum system based on the SIRMs dynamically connected fuzzy inference model

Abstract A new fuzzy controller is presented based on the single input rule modules (SIRMs) dynamically connected fuzzy inference model for upswing and stabilization control of inverted pendulum system. The fuzzy controller takes the angle and angular velocity of the pendulum and the position and velocity of the cart as its input items, and the driving force as its output item. Each input item is assigned with a SIRM and a dynamic importance degree. When the pendulum locates at the pending domain, the fuzzy controller becomes an upswing controller by using the saturation feature of the membership functions of the pendulum angle. When the pendulum locates at the upright domain, the fuzzy controller then becomes a stabilization controller and realizes smoothly the pendulum angular control and the cart position control in parallel by using the SIRMs and the dynamic importance degrees. The fuzzy controller has a simple structure and is easily understandable compared with the other approaches. Simulation results show that the fuzzy controller can swing up the pendulum from the pending position and then stabilize the whole system in about 3.0 s.

A fuzzy Actor-Critic reinforcement learning network

One of the difficulties encountered in the application of reinforcement learning methods to real-world problems is their limited ability to cope with large-scale or continuous spaces. In order to solve the curse of the dimensionality problem, resulting from making continuous state or action spaces discrete, a new fuzzy Actor-Critic reinforcement learning network (FACRLN) based on a fuzzy radial basis function (FRBF) neural network is proposed. The architecture of FACRLN is realized by a four-layer FRBF neural network that is used to approximate both the action value function of the Actor and the state value function of the Critic simultaneously. The Actor and the Critic networks share the input, rule and normalized layers of the FRBF network, which can reduce the demands for storage space from the learning system and avoid repeated computations for the outputs of the rule units. Moreover, the FRBF network is able to adjust its structure and parameters in an adaptive way with a novel self-organizing approach according to the complexity of the task and the progress in learning, which ensures an economic size of the network. Experimental studies concerning a cart-pole balancing control illustrate the performance and applicability of the proposed FACRLN.

Stabilization fuzzy control of parallel-type double inverted pendulum system

A fuzzy controller for stabilizing parallel-type double inverted pendulum system is presented, based on the single input rule modules (SIRMs) dynamically connected fuzzy inference model. By using the SIRMs and the dynamic importance degrees, the angular controls of the two pendulums and the position control of the cart are done entirely in parallel and the priority orders of the three controls are automatically adjusted according to control situations. Simulation results show that the fuzzy controller with a simple and intuitive structure can stabilize completely a parallel-type double inverted pendulum system within 10 seconds. This is the first result for a fuzzy controller to realize complete stabilization of a parallel-type double inverted pendulum system.