A novel multi-sensor multiple model particle filter with correlated noises for maneuvering target tracking

Aiming at the effective realization of particle filter for maneuvering target tracking in multi-sensor measurements,a novel multi-sensor multiple model particle filtering algorithm with correlated noises is proposed.Combined with the kinetic evolution equation of target state,a multi-sensor multiple model particle filter is firstly constructed,which is also used as the basic framework of a new algorithm.In the new algorithm,in order to weaken the adverse influence from random measurement noises in the measuring process of particle weight,a weight optimization strategy is introduced to improve the reliability and stability of particle weight.In addition,considering the correlated noise existing in the practical engineering,a decoupling method of correlated noise is given by the rearrangement and transformation of the state transition equation and measurement equation.Since the weight optimization strategy and noise decoupling method adopt respectively the center fusion structure and the off-line way,it improves the adverse effect effectively on computational complexity for increasing state dimension and sensor number.Finally,the theoretical analysis and experimental results show the feasibility and efficiency of the proposed algorithm.     

Two-stage prediction and update particle filtering algorithm based on particle weight optimization in multi-sensor observation

The reasonable measuring of particle weight and effective sampling of particle state are consid- ered as two important aspects to obtain better estimation precision in particle filter. Aiming at the comprehensive treatment of above problems, a novel two-stage prediction and update particle filte- ring algorithm based on particle weight optimization in multi-sensor observation is proposed. Firstly, combined with the construction of muhi-senor observation likelihood function and the weight fusion principle, a new particle weight optimization strategy in multi-sensor observation is presented, and the reliability and stability of particle weight are improved by decreasing weight variance. In addi- tion, according to the prediction and update mechanism of particle filter and unscented Kalman fil- ter, a new realization of particle filter with two-stage prediction and update is given. The filter gain containing the latest observation information is used to directly optimize state estimation in the frame- work, which avoids a large calculation amount and the lack of universality in proposal distribution optimization way. The theoretical analysis and experimental results show the feasibility and efficiency of the proposed algorithm.     

Robust SLAM using square-root cubature Kalman filter and Huber’ s GM-estimator

Mobile robot systems performing simultaneous localization and mapping ( SLAM) are generally plagued by non-Gaussian noise.To improve both accuracy and robustness under non-Gaussian meas-urement noise, a robust SLAM algorithm is proposed.It is based on the square-root cubature Kal-man filter equipped with a Huber’ s generalized maximum likelihood estimator ( GM-estimator) .In particular, the square-root cubature rule is applied to propagate the robot state vector and covariance matrix in the time update, the measurement update and the new landmark initialization stages of the SLAM.Moreover, gain weight matrices with respect to the measurement residuals are calculated by utilizing Huber’ s technique in the measurement update step.The measurement outliers are sup-pressed by lower Kalman gains as merging into the system.The proposed algorithm can achieve bet-ter performance under the condition of non-Gaussian measurement noise in comparison with benchmark algorithms.The simulation results demonstrate the advantages of the proposed SLAM algorithm.     

Passive Target Tracking Based on Current Statistical Model

Bearing-only passive tracking is regarded as a nonlinear hard tracking problem. There are still no completely good solutions to this problem until now. Based on current statistical model, the novel solution to this problem utilizing particle filter （PF） and the unscented Kalman filter （UKF） is proposed. The new solution adopts data fusion from two observers to increase the observability of passive tracking. It applies the residual resampling step to reduce the degeneracy of PF and it introduces the Markov Chain Monte Carlo methods （MCMC） to reduce the effect of the “sample impoverish”. Based on current statistical model, the EKF, the UKF and particle filter with various proposal distributions are compared in the passive tracking experiments with two observers. The simulation results demonstrate the good performance of the proposed new filtering methods with the novel techniques.     

Improved hand tracking algorithm in video sequences for intelligent rehabilitation

Intelligent rehabilitation system is an active research topic. It is motivated by the increased number of limb disabled patients. Human motion tracking is the key technology of intelligent rehabilitation system, because the movement of limb disabled patients needs to be localized and learned so that any undesired motion behavior can be corrected in order to reach an expectation. This paper introduces a realtime tracking system of human hand motion, specifically intent to be used for home rehabilitation. Vision sensor (camera) is employed in this system to track the hand movement, and the improved Camshift algorithm and Kalman filter are used to implement dynamic hand tracking in the video. CAMSHIFT algorithm is able to track any kind of target colors by building a histogram distribution of the H channel in HSV color space from the region of interests selected by users at the initial stage. Kalman filter is able to predict hand location in one image frame based on its location data detected in the previous frame. The experimental results show that this system can track 2D hand motion and has acceptable accuracy by using the two algorithms properly. The new algorithm proposed in this paper can not only deal with the skin color interference problems, but also deal well with the track of complex background.     

Trajectory tracking control of quantum systems

For quantum state trajectory tracking of density matrix in Liouville equation of quantum systems,with the help of concept in quantum system control,one can apply unitary transformation both to controlled system and free-evolutionary target system such as to change the time-variant and non-stationary target system into a stationary state.Therefore,the quantum state trajectory tracking problem becomes a steering one.State steering control law of the system transformed is designed by means of the Lyapunov stability theorem.Finally,numerical simulation experiments are given for a five-level energy quantum system.The comparison analysis of original system’s trajectory tracking with other method illustrates the advantage in control time of the method proposed.     

Multiscale Multisensor Data Fusion and Application in High Precision Marking and Cutting Robot System

The multisensor online measure system for high precision marking and cutting robot system is designed and the data fusion method is introduced,which combines augment state multiscale process with extend Kalman filter.The technology measuring the three-dimensional deforming informatioin of profiled bars is applied.The experimental result shows that applying the multisensor data fusion technology can enhance the measure precisioin and the reliability of measure system.     

Pose measurement using quaternion and Kalman filter

In this paper, a method combining perspective-n-point （PnP） and novel iteration algorithm is developed to measure the pose of a target in high precision for Tele-LightSaber game. The PnP algorithm is used to obtain a rough pose, which is taken as the initial value of the iteration algorithm. The iteration algorithm utilizes the unit quaternions to represent the rotations. Then the result is optimized with Kalman filter. Considering the real-time and accuracy of the pose measurement, a fast feature extraction algorithm including object location, edge detection and corner detection is adopted to get the corners in high precision. The experiments and results verify the effectiveness of the proposed method.     

A Benefit Measurement Framework for an Online Contract Change Management System

The contract change management (CCM) system is an online collaboration tool, which supports the contract change management process of new engineering contract (NEC) contract. It is currently used in practice on a wide range of civil, power, and building projects in the UK. The aim of this study is first to identify the tangible and intangible benefits of the system through a user survey; then to develop a framework to measure these benefits. A questionnaire survey is conducted, which reveals that CCM can help project teams improve process effectiveness, reduce risk of process failures, and increase the compliance with NEC requirements. Subsequently, a spreadsheet tool is developed to allow CCM users to calculate the tangible cost savings as a result of using the system. For intangible benefits, a change management maturity model (CM3) is developed, which provides a measurement framework for assessing the improvement of a project team’s capability in dealing with contract changes.     

A Particle Filter of Blind Equalization and Multiuser Detection in Asynchronous DS/CDMA Systems

The particle filter （PF） is proposed to be the asynchronous direct-sequence code-division multiple-access （DS/CDMA） multiuser detector without knowing the channel state information. The PF performs symbol detection according to the joint posterior density probability of simulated particles including relative delays, fading gains and symbols via sequential importance sample and resample. A simplified scheme is also proposed by separating the indepent relative delays and fading with symbols. These parameters are modeled as the extended aggressive processes and estimated by the Kalman filter, so as to provide their arbitrary distribution for symbol detection. Simulation results show that the bit error rate of the PF is less than conventional detectors. Moreover, the complexity of PF is moderate comparable to other nonlinear suboptimal approaches.     

Autonomous control of multi-fingered hand

This paper describes a novel autonomous control strategy of multi-fingered hand based on a modular control system of dexterous manipulation. A simple proportional-integral-derivative(PID) position control with friction compensation, which requires few friction parameters, is used to realize accurate and smooth trajectory tracking in pregrasp phase. In grasp and manipulation phases, an event-driven switcher is adopted to determine the switching between unconstrained position control and constrained torque control, and an improved explicit integral force control strategy is implemented to realize simultaneously stable contact transition and accurate force tracking. Experimental results have verified the effectiveness of the proposed autonomous control strategy of multi-fingered hand.     

A probabilistic paradigm for handling uncertain objects in GIS by randomized graph algebra

Probability theory faces difficulties when it is applied to describing uncertain objects in geographic information system (GIS). This is mainly due to the fact that an object in GIS is normally described by a series of discrete vertexes. Modeling uncertainty objects should be therefore based on error of the composed vertexes. This type of model is normally complex and relatively difficult to implement because of many unknown factors, such as the number of vertexes of a polygon, error nature of each individual vertex and error correlation among the vertexes. In this paper, a probabilistic paradigm for handling uncertain objects in GIS by randomized graph algebra is presented. The theoretical basis for this paradigm is the randomized graph algebra-a probability theory for graph-which is newly proposed in this study. Classical probability theory is based on numerical algebra and is also an extension of numerical algebra by further defining probability density within a numerical domain. In the same token, this study begins with defining graph algebra as the basis for probability theory for graph. First, we adopt the theory of graph algebra and further refine the theory by defining the modulo operation for graph. As a result, a graph can thereafter be treated as a "number" and operated by "addition", "subtraction" and others. Second, we construct a measure space by generating sigma-algebra and defining measurable function upon it. The measure space becomes a probability space when the measurable function is a probability density function. Third, we propose the probabilistic paradigm for describing and inferring the uncertainty of geometric objects in GIS by applying the developed randomized graph algebra.     

Motion Geometric Active Contours： Tracking Nonrigid Objects in Clutter Background

MGAC (Motion Geometric Active Contours), a new variational framework of geometric active contours to track multiple nonrigid moving object‘s in the clutter background in image sequences is presented. This framework, incorporating with the motion edge information, consists of motion detection and tracking stages. At the motion detection stage, the motion edge map provides an approximate edge map of the moving objects. Then, a tracking stage, merely using thestatic edge information, is considered to improve the motion detection result. Force field regularization method is used to extend the capture range of the edge attraction force field in both stages.Experiments demonstrate that the proposed framework is valid for tracking multiple nonrigid objects in the clutter background.     

Robust direct adaptive fuzzy control for nonlinear MIMO systems

For a class of nonlinear multi-input multi-output systems with uncertainty, a robust direct adaptive fuzzy control scheme was proposed. The feedback control law and adaptive law for parameters were derived based on Lyapunov design approach. The overall control scheme can guarantee that the tracking error converges in the small neighborhood of origin, and all signals of the closed-loop system are uniformly bounded. The main advantage of the proposed control scheme is that in each subsystem only one parameter vector needs to be adjusted on-line in the adaptive mechanism, and so the on-line computing burden is reduced. In addition, the proposed control scheme is a smooth control with no chattering phenomena. A simulation example was proposed to demonstrate the effectiveness of the proposed control algorithm.     

Real-time vehicle tracking for traffic monitoring systems

A real-time vehicle tracking method is proposed for traffic monitoring system at road intersec-tions, and the vehicle tracking module consists of an initialization stage and a tracking stage .Li-cense plate location based on edge density and color analysis is used to detect the license plate re -gion for tracking initialization .In the tracking stage , covariance matching is employed to track the license plate .Genetic algorithm is used to reduce the computational cost .Real-time image tracking of multi-lane vehicles is achieved .In the experiment , test videos are recorded in advance by record-ers of actual E-police systems at several different city intersections .In the tracking module , the av-erage false detection rate and missed plates rate are 1.19%, and 1.72%, respectively.     

A novel extended Kalman filter for a class of nonlinear systems

Estimation of the state variables of nonlinear systems is one of the fundamental and significant problems in control and signal processing. A new extended Kalman filtering approach for a class of nonlinear discrete-time systems in engineering is presented in this paper. In contrast to the celebrated extended Kalman filter (EKF), there is no linearization operation in the design procedure of the filter, and the parameters of the filter are obtained through minimizing a proper upper bound of the mean-square estimation error. Simulation results show that this filter can provide higher estimation precision than that provided by the EKF.     

Generalized Reactive Manufacturing System

Generalized reactive manufacturing system named GRMS is introduced. GRMS is a human-centered system based on Multi-agent. Its management and control organization is made up of three types of agents named device agent, task agent and shop-floor agent. GRMS adopts a top-down and bottom-up competition and cooperation strategy based on the dynamic sifter and funnel . To constrain the behavior of agents, a reward and penalty policy is introduced into the system and the closed-loop adjustment of GRMS is realized through such policy. Agents for the same task should be cooperated with each other and agents for different tasks should compete for survival in the dynamic changing environment. A distributed-hierarchical architecture with three levels of master-slave relationships among agents are proposed. Self-propelled process planning is also discussed. In order to evaluate GRMS, a time-driven simulation system-GRMOSS is developed to check the physical consistency of GRMS.     

Real-time filter based on the quasi-accurate detection of gross errors

Because of the sensitivity of the Kalman framework to gross errors, proper techniques for detection of gross errors are necessary. By integrating the selection of quasi-accurate observations and the Kalman framework, a new filter called the quasi-accurate filter (QUAF) is developed. The expansibility and implementation scheme of the new algorithm are then discussed in detail, and the reliability matrix for the Kalman filter is proposed to analyze the reliability of the filters with different detection technologies. Finally, the experimental results from a real world case study are used to validate the conclusions. The QUAF carries out the preliminary selection of the quasi-accurate observations (QAOs) using the innovation of the Kalman filter, and use the check QAOs to determine reasonable observations. This causes the QUAF to handle more easily and possess wider expansibility. QUAF can be reformulated to the special cases of several common detection methods, such as the innovation method, robust estimation and quasi-accurate detection (QUAD). Since only reasonable observations are used, the QUAF has better detection accuracy and stronger avoidance of gross errors than the innovation method and robust estimation. Meanwhile, compared with QUAD methods, QUAF introduces the state-predicted model, requiring fewer quasi-accurate observations and making it more suitable for systems with complicated observation structures or sparse observations.     

Attitude Determination for MAVs Using a Kalman Filter

This paper presents a Kalman filter to effectively and economically determine the Euler angles for micro aerial vehicles （MAVs）, whose size and payload are severely limited. The filter uses data from a series of micro-electro mechanical system sensors to determine the selected 3 variables of the direction cosine matrix and the bias of the rate gyro sensors as state elements in a dynamic model, with the gravitational acceleration to build a measurement model. For high speed maneuvers, rigid motion equations are used to correct the measurements of the gravitational acceleration. The filter is designed to automatically tune its gain based on the dynamic system state. Simulations indicate that the Euler angles can be determined with standard deviations less than 3°. The algorithm was successfully implemented in a miniature attitude measurement system suitable for MAVs. Aerobatic flights show that the attitude determination algorithm works effectively. The attitude determination algorithm is effective and economical, and can also be applied to bionic robofishs and land vehicles, whose size and payload are also greatly limited.