Traction motor systems for electric vehicles provides fast torque response that may let the driver input result in excessive propelling or regenerative braking torque, which can reduce tire-road adherence. The same situation can also occur due to excessive wheel slip that tends to spin or skid the wheels, resulting in safety issues or energy losses. However, to measure the slip ratios for vehicle control, vehicle speed measurement instruments are needed, which are too expensive to implement on commercial electric vehicles. This paper presents a slip ratio observer based on a Kalman-like observer for a state-affine system, which does not use the tire-road friction model and does not require knowledge of vehicle parameters. We also propose a wheel slip controller, based on fuzzy sliding mode control (FSMC), which uses a slip ratio estimate given by the observer. The FSMC is capable of manipulating the slip ratio to at a desired value, for which the traction or braking force is maximum. The slip ratio observer and FSMC have been tested extensively with numerical simulations, and the results confirm their validity.

This paper proposes a novel adaptive cruise control (ACC) system for vehicles that addresses the issues of fuel economy, tracking capability, and driver desired response. The controller is synthesized by applying an adaptive optimal control (ACC) algorithm. A cost function is developed that considers the conflicting objectives of high fuel economy, minimizing tracking error, and providing a driver desired response. The linear constraints are transformed to another cost function using a penalty function method. A vehicle speed estimator was developed to provide the longitudinal vehicle speed information to the ACC system. Simulation results show that the proposed ACC system performs provides good tracking capability and fuel economy, and the desired driver response.

This paper presents a target-surround feature attention (TSFA) model for constructing attention-based visual tracking algorithm. This model extracts attentive region by distinguishing the color contrast between the interested target and its surround. A preference generator provides online feature transformation to update the target/surround biasing masks that describes the color composition associated with the target and its surround. Output of the TSFA model is a saliency map representing occurrence possibility of the target. Tracker based on the mean shift algorithm is used to lock and locate the target on the saliency map. Experimental results show that visual tacking algorithm with the TSFA model may adapt to noisy images under changing illumination.

Computational visual attention (CVA) model is one of the methods which focus on finding region of interesting (ROI) in an image or in a scene. Similarity attention is one important task in CVA. If there are many objects in a scene, people will pick up the most abnormal one, which perhaps the similar one or dissimilar one, according to the composition objects of the scene. Capability of similarity attention enables human vision to promptly focus on similar or dissimilar regions in a scene. This paper implements this capability in the CVA model by attaching a high-level similarity comparison function to find ROI in the scene. The output of the model simulates the serial search mode and more approach to human visual behavior. Experimental results show that the function of similarity attention can be achieved successfully.

Automatic train regulation (ATR) dominates the service quality, transport capacity, and energy usage of a metro-line operation. The train regulator aims to maximize the schedule/headway adherence while minimizing the energy consumption. This paper presents a traffic-energy model to characterize the complicated dynamics with regard to the traffic and the energy consumption of a metro line, and devises an adaptive-optimal-control (AOC) algorithm to optimize the train regulator through reinforcement learning. The updating rules for reinforcement learning are deduced from the discrete minimum principle. Testing with field traffic data, the AOC algorithm succeeds in the optimization of the train regulator; no matter the system is disturbed by passenger-flow fluctuations or by frequently minor delays. The results also show that better train regulation with less energy consumption is attainable through the running-time and dwell-time controls.

Automatic train regulation, which is a core function of the signaling system, concerns the headway/schedule adherence that dominates the transport capacity and punctuality of a metro line. The main difficulty in synthesizing a traffic regulator is that an accurate traffic model is inaccessible. This paper presents an adaptive optimal control (AOC) algorithm that can approximate the optimal traffic regulator by learning traffic data with artificial neural networks. The AOC algorithm is derived from the discrete minimum principle and organized in the critic-actor architecture of reinforcement learning to carry out sequential optimization forward in time. The critic network receives no signal from the traffic model so that the prediction of the future cost and the optimization of the traffic regulator are not biased by modeling errors. The efficacy of the AOC algorithm in the traffic regulation is verified in a simulated system using traffic data acquired from a real metro line.

Communication based train control bring the possibility of moving block control of metro line. The system's transport capacity and punctuality in service depends on the capability of automatic train regulation. The automatic train regulation problem is essentially nonlinear and stochastic, and limited by operating hours of metro line. Neuro-dynamic programming with recurrent critic is shown able to find near-optimal solution more rapidly and accurately than that with forward critic design. Multilayered perceptrons and back-propagation technique are used to construct the recurrent critic and other components associated with adaptive critic design. Comparisons of automatic train regulation referring to Taipei metro data are made for recurrent critic regulator, forward critic regulator, and linear quadratic regulator. The neuro-dynamic programming with recurrent critic design is shown robust with respect to modeling error and excellent in convergence.

Because of the disturbance of operation environment in mass rapid transit (MRT) system, the robustness against disturbance and the schedule punctuality under control constraint are important issues to be considered in designing Automatic Train Regulation (ATR) for MRT system. In this paper, the study on suitable traffic model for designing ATR system and ATR design based on adaptive critic design (ACD) of approximated dynamic programming, specifically on dual heuristic programming (DHP) are presented. Moreover, the method to deal with control constraint and applying gain scheduling to deal with the time variant environment of MRT system are addressed as well. For comparison, simulations with real operation environment data are done for ATRs designed by both adaptive critic method and LQ optimization method.

Two dimensional shape recognition needs to remove the effect of geometric transformations from the contour pattern. This is usually obtained by transforming the contour pattern into an affine invariant form. In this paper, the cosine transform is used to decompose the contour pattern of the object shape into a set of synthesized feature signals (SFS). Then the SFSs are used to evaluate the cosine transform based synthesized affine invariant function (CSAIF) for the shape recognition. The information loss of representing the contour pattern with the CSAIF is minimized by simply neglecting the translation component. This assures high accuracy in recognizing the object shape by discriminating its CSAIF from others. Experiments examine the performance and robustness of the CSAIF based shape recognition and the results are compared with that of using the Fourier affine invariant function.

This paper presents an adaptive critic neuro-fuzzy control design, which enables learning from scratch to achieve the control objective. The learning algorithm is derived from the Dual Heuristic Programming (DHP) method to approximate optimal control. The learning structure contains the action, critic and verification neuro-fuzzy networks each corresponding to the first-order Sugeno fuzzy model. The critic and verification networks together forms a reinforcement learning scheme to estimate the future derivative Bellman function. The transparency of the network sensitivity function makes the update rule of the action network simple. Simulation results show that the adaptive critic neuro-fuzzy control system enables the motorized vehicle with simply two coaxial wheels learning from scratch to achieve velocity control while keeps the intermediate body in balance.

Human attention behavior relies not only on the natural contrastive information but also much top-down information in the mind. The intention represents preference or repugnance for specific image property undoubtedly affects the visual attention. By introducing the intentional features and the fuzzy inference system into the computational model, the Intention-oriented Computational Visual Attention (ICVA) is developed and shown able to combine the intention with the optical excitation to And the attentive spots. An intention about favorite colors, a desired object or even multiple desired objects is expressed as a combination of preference vectors and topological relationships. Referring to the preference vectors, the intentional features are extracted. According to the topological relationships, the linguistic rules of the fuzzy inference system are built to infer the results. Several experiments have been conducted and the results verify the feasibility of the proposed design.

This paper will address the problem of controlling an unknown MIMO nonlinear affined system. A fuzzy sliding mode controller (FSMC) is used to approximate the equivalent control by using an on-line fuzzy adaptation scheme, and then the hitting control is appended to show that the proposed FSMC can result in a closed-loop system, which is stable. This scheme also provides the designers flexibility to design and to implement fuzzy rule base without domain experts and without mathematical model. The robust adaptive scheme is shown to be able to guarantee that the output tracking error can converge to, a residual set ultimately by a two-dimensional inverted pendulum system.

A robust fuzzy model-following control system is proposed for the control of robot manipulators. The application field to n-link robot manipulators with torque disturbance and measurement noise is addressed. The control objective is obtained by tailoring a nominal adaptation process of parameters to implement appropriate function approximation and facilitating a self-tuning mechanism on the consequent membership functions to overcome the equivalent uncertainty. A novel fuzzy system with self-tuning mechanism provides robust property and the rule-base in the form of "IF situation THEN the control input". The proposed multilayer fuzzy logic controller can improve both transient and stability margins without a priori knowledge about the dynamic model or parameters of the robotic system. Using the Lyapunov stability method, the uniform ultimate boundedness of tracking error has been proved. The performance is demonstrated by simulating the control of a two-link robot in various situations.

In this paper, we show how an active binocular head, the IIS head, can be easily calibrated with very high accuracy. Our calibration method can also be applied to many other binocular heads. In addition to the proposal and demonstration of a four-stage calibration process, there are three major contributions in this paper. First, we propose a motorized-focus lens (MFL) camera model which assumes constant nominal extrinsic parameters. The advantage of having constant extrinsic parameters is to having a simple head/eye relation. Second, a calibration method for the MFL camera model is proposed in this paper, which separates estimation of the image center and effective focal length from estimation of the camera orientation and position. This separation has been proved to be crucial; otherwise, estimates of camera parameters would be very noise-sensitive. Thirdly, we show that, once the parameters of the MFL camera model is calibrated, a nonlinear recursive least-square estimator can be used to refine all the 35 kinematic parameters. Real experiments have shown that the proposed method can achieve accuracy of one pixel prediction error and 0.2 pixel epipolar error, even when all the joints, including the left and right focus motors, are moved simultaneously. This accuracy is good enough for many 3D vision applications, such as navigation, object tracking and reconstruction.

This paper proposes a new closed-form solution for identifying the kinematic parameters of an active binocular head having four revolute joints and two prismatic joints by using three-dimensional (3-D) point (position) measurements of a calibration point. Since this binocular head is composed of off-the-shelf components, its kinematic parameters are unknown. Therefore, we can not directly apply those existing nonlinear optimization methods. Even if we want to use the nonlinear optimization methods, a closed-form solution can be first applied to obtain accurate enough initial values. Hence, this paper considers only methods that provide closed-form solutions, i.e., those requiring no initial estimates. Notice that most existing closed-form solutions require pose (i.e., both position and orientation) measurements. However, as far as we know, there is no inexpensive technique which can provide accurate pose measurements. Therefore, existing closed-form solutions based on pose measurements can not give us the required accuracy. As a result, we have developed a new method that does not require orientation measurements and can use only the position measurements of a calibration point to obtain highly accurate estimates of kinematic parameters using closed-form solutions. The proposed method is based on the complete and parametrically continuous (CPC) kinematic model, and can be applied to any kind of kinematic parameter identification problems with or without multiple end-effecters, providing that the links are rigid, the joints are either revolute or prismatic and no closed-loop kinematic chain is included.

In camera calibration, due to the correlations between certain camera parameters, an estimate of a set of camera parameters which minimizes a given criterion does not guarantee that the estimates of the physical camera parameter are themselves accurate. This problem has not drawn much attention from our computer vision society because conventional computer vision applications require only accurate 3D measurements and do not care much about the values of the physical parameters as long as their composite effect is satisfactory. However, when dealing with an active vision system, accuracy of the physical parameters is very critical because we need that accuracy to establish the relation between the motor positions and the camera parameters (both intrinsic and extrinsic). The contribution of this work is mainly in error analysis of camera calibration, especially in the accuracy of the physical camera parameters themselves, for four different types of calibration problems. With our error analysis, the most suitable camera calibration technique and calibration configuration for providing accurate camera parameters can be determined. Based on this error analysis, we have developed a method for calibrating our eight-degree-of-freedom binocular system. Real experiments have shown that this method can achieve accuracy of one pixel prediction error and 0.2 pixel epipolar error, even when all the eight joints, including the focus motors, are moved simultaneously. This high accuracy greatly alleviates the difficulties encountered in applying the active vision approach to automatic reconstruction of 3D objects and environments, which is our current endeavor.

This paper presents an adaptive fuzzy computed-torque controller, that enhances fuzzy controllers with an embedded adjustable two-stages credit assignment and self-learning capability, for uncertain robots, to online track a prescribed trajectory. An adaptation law for the parameters of controller is combined with the dead-zone technique to guarantee a given attenuation region of tracking error in the presence of torque disturbance. Simulations of a two-link robot carrying a heavy load illustrate the effectiveness and attenuation capability of the controller for online trajectory tracking in the presence of inertial parameters uncertainties and torque disturbances.

This paper proposes a new closed-form solution for identifying the kinematic parameters of an active binocular head by using a single calibration point. This method is based on the complete and parametrically continuous (CPC) kinematic model, and can be applied to any kind of kinematic parameter identification problems with or without multiple end-effectors, providing that the links are rigid, the joints are either revolute or prismatic and no closed-loop kinematic chain is included. As a practical example, this paper focuses on the calibration of a binocular head having four revolute joints and two prismatic joints. Simulation and real experiments have shown that the proposed method of using point measurements can achieve much higher accuracy than that of using pose measurements.

The paper by Zhuang and Roth (ibid. vol.9, p.174-85 (1993)) presents a linear solution to kinematic parameter identification of robot manipulators. With their method, the orientation parameters for all revolute joints have to be solved first before solving for the translation parameters for all revolute joints and the orientation parameters for all prismatic joints simultaneously. Our major modification here is to decompose the kinematic parameters estimation problem into many subproblems of a single joint axis such that the complexity is reduced and easier implementation is derived. In addition, this gives a unified solution for any serial manipulator with an arbitrary combination of prismatic and revolute joints.>

The authors investigate the effect of neglecting lens distortion, and present a theoretical analysis of the calibration accuracy. The error bound derived is a function of a few factors, including the number of calibration points, the observation error of 2-D image points, the radial lens distortion coefficient, and the image size and resolution. This error bound provides a guideline for selecting both a proper camera calibration configuration and an appropriate camera model while providing the desired accuracy. Experimental results from both computer simulations and real experiments are given.>