Interactive Multi-model Research Articles

Joint multi-Gaussian mixture model and its application to multi-model multi-bernoulli filter

The existing multi-model Gaussian mixture-multi-Bernoulli (MM-GM-MB) filter lacks the ability to maintain parallel state estimates under different models. As a result, the change of model-related likelihoods lags behind target maneuvers. To address this problem, a joint multi-Gaussian mixture MB (JMGM-MB) filter is proposed. We first propose a JMGM model. Each single-target state estimate is expressed as a set of parallel model-related Gaussian functions with model probabilities, and a weight characterizes the probability of this state estimate. Based on the Bayesian rule, we derive the updating methods of the weights, model probabilities and model-related means, covariances that are collectively called JMGM components. Then, we approximate the state densities of MB filter using the JMGM model. Based on the interactive multi-model (IMM) method, we derive the interacting, prediction and estimation methods of JMGM components. Then, we provide matched association and merging methods for JMGM components. In nonlinear tracking scenarios where sensors simultaneously carry out translations and rotations, we derive a method based on the derivative rule for composite functions to compute the linearized observation matrix. Simulations demonstrate that whether in active or passive tracking, the proposed JMGM-MB filter overcomes the likelihood lag problem and owns higher accuracy of uncertain maneuvering target tracking than the existing MM-GM-MB filter, requiring lower computational expense.

Retracted: Application of Improved Interactive Multimodel Algorithm in Player Trajectory Feature Matching

Retracted: Application of Improved Interactive Multimodel Algorithm in Player Trajectory Feature Matching

A multimodal fusion positioning method with adaptive driving behavior for intelligent vehicles

Due to the difficulty of Global Navigation Satellite System (GNSS) in urban canyon environments, GNSS and inertial navigation system (INS) fusion positioning is widely used in the field of intelligent driving. In actual positioning, INS usually uses the vehicle physical model as a constraint to calculate inertial measurement unit (IMU) data to supplement the vehicle position and attitude as GNSS positioning information. However, the existing algorithms usually use a fixed physical model. When the vehicle’s driving behavior changes, such as lane-changing, acceleration, deceleration, etc., the fixed model cannot provide long-term constraints on cumulative errors, resulting in a decrease in fusion positioning accuracy and stability. Therefore, this paper proposes a multimodal fusion positioning method with adaptive driving behavior for intelligent vehicles. First, to adapt to multimodal driving behavior, a multimodal assisted positioning physical model with adaptive driving behavior (MAPM) is established, which expresses different physical models probabilistically based on driving behavior and historical data. Secondly, high-precision estimation of vehicle positioning parameters is achieved based on model probability and interactive multi-model (IMM). Finally, effective fusion of MAPM/GNSS/INS positioning information is achieved under the federated Kalman filtering framework. The simulation results show that compared to traditional GNSS/INS fusion positioning, the proposed method can effectively improve positioning accuracy and stability, especially in driving conditions with changes in driving behavior.

Lane change trajectory prediction considering driving style uncertainty for autonomous vehicles

Lane change trajectory prediction is crucial for autonomous vehicles (AVs) to assess their own driving safety in advance. However, there are significant uncertainties in the implementation of such prediction, including different behaviors caused by agent–agent interaction and driving styles of drivers. While prototype trajectories can serve as a means to represent typical motion patterns and enhance trajectory prediction performance, their utilization in modeling motion patterns tends to overlook the influence of agent–agent interactions and vehicle dynamics. This paper proposes a fusion algorithm that considers driving style and vehicle dynamics to address these uncertainties. The algorithm involves a long short-term memory (LSTM) lane change behavior recognition model that mines key features of agent–agent interaction through the attention mechanism. The Gaussian process (GP) motion modeling trajectory prediction (GPMM-TP) algorithm considers the driving style of the prototype trajectory based on the behavior recognition results. To further improve short-term and long-term prediction, the interactive multi-model (IMM) algorithm is used to assign probability weights to the GP model and the Extended Kalman Filter (EKF) model based on prediction accuracy, taking into account the driving styles and the vehicle dynamics. The proposed algorithm provides a promising approach to improving the accuracy of lane change trajectory prediction for AVs, and its effectiveness is demonstrated using the HighD dataset.

A multi-sensor cooperative detection target tracking method based on radar-optical linkage control

Radar is a common means of tracking a target, and with active enemy interference, it often causes the target to lose its track, thus causing the radar to lose continuous tracking of the target. To improve the tracking effect, a multi-sensor cooperative detection target tracking method based on radar photoelectric linkage control was established. The study is based on radar photoelectric linkage, constant velocity (CV), constant acceleration (CA) and current statistical model (CSM) as the mathematical model of moving targets for this study. Improved interactive multi-model (IMM) and standard IMM were compared for targets in different motion states, as well as single sensor electronic support measures (ESM) and multi-sensor electronic support measures (ESM), infrared search and track (IRST). The research results show that in variable speed motion, the improved IMM algorithm and multiple sensors are used for target tracking. The azimuth and elevation tracking errors of the target are low, which can effectively solve the problem of model mismatch during the conversion of motion modes such as CV and CA. The azimuth and elevation image curves fluctuate smoothly, and have high stability. This method can achieve better tracking results.

Interactive multi-model fault diagnosis method of switched reluctance motor based on low delay anti-interference.

Given fault false alarm and fault control failure caused by the decrease of fault identification accuracy and fault delay of Switched Reluctance Motor (SRM) power converter in complex working conditions, a method based on the Interactive Multi-Model (IMM) algorithm was proposed in this paper. Besides, the corresponding equivalent circuit models were established according to the different working states of the SRM power converter. The Kalman filter was employed to estimate the state of the model, and the fault detection and location were realized depending on the residual signal. Additionally, a transition probability correction function of the IMM was constructed using the difference of the n-th order to suppress the influence of external disturbance on the fault diagnosis accuracy. Concurrently, a model jump threshold was introduced to reduce delay when the matched model was switched, so as to realize the rapid separation of faults and effective fault control. The simulation and experiment results demonstrate that the IMM algorithm based on low delay anti-interference can effectively reduce the influence of complex working conditions, improve the anti-interference ability of SRM power converter fault diagnosis, and identify fault information accurately and quickly.

Orbit Determination and Thrust Estimation for Noncooperative Target Using Angle-Only Measurement

The classical interactive multimodel (IMM) algorithm has some disadvantages in tracking a noncooperative continuous thrust maneuvering spacecraft, such as poor steady-state accuracy, difficult selection of subfilter parameters, and mismatched model jump. To address the abovementioned problems, a variable-dimensional adaptive IMM strong tracking filtering algorithm (VAIMM-STEKF) is proposed to estimate the spacecraft’s position, velocity, and maneuvering acceleration state. VAIMM-STEKF contains 2 models, model 1 and model 2, which correspond to the tracking of the spacecraft in maneuvering and nonmaneuvering situations. Model 1 estimates the position and velocity of the spacecraft to ensure tracking accuracy when no maneuver occurs. Model 2 is a strong tracking filter with an augmented state. The adaptive IMM algorithm adjusts the fixed Markov transfer matrix in real time according to the model output probability. According to the different states of the spacecraft, the corresponding model interactive fusion method, together with the strong tracking filter, is adopted to ensure fast tracking when the spacecraft state changes. This method can also adapt to continuous thrust maneuvering spacecraft with different orders of magnitude. Simulation results show that the position accuracy of VAIMM-STEKF can be improved by approximately 27% and the speed accuracy can be enhanced by approximately 17% under different levels of maneuvering acceleration compared with those of the IMM algorithm. The convergence speed of VAIMM-STEKF is also better than the IMM algorithm.

Слежения надводных маневрирующих целей на основе интерактивного многомодельного алгоритма

The volume of sea and river transportation is increasing every year. As a result, the requirements for security systems in water transport are increasing. Among the measures to improve the safety of navigation, one of the most important ones is the improvement of data processing methods in ship control systems, including the target trajectory tracking. In recent years, much attention has been paid to the target trajectory tracking. An interactive multi-model (IMM) algorithm is proposed for tracking maneuvering surface vessels. The simulation of the work of the algorithm for trajectory filtering of targets in the Matlab environment is given. The next direction is suggested in the conclusion.

Space Noncooperative Target Trajectory Tracking Based on Maneuvering Parameter Estimation

The space noncooperative target maneuvering trajectory tracking is essential for the safety of the on-orbit spacecraft. For the noncooperative target, the maneuvering model is complex and changeable. Besides, the maneuvering dynamics model, the operation time, and the maneuvering frequency are previously unknown. It is difficult to achieve high-precision maneuvering trajectory tracking. In this paper, a novel real-time maneuvering trajectory tracking algorithm is developed, in which the maneuvering trajectory of the noncooperative target is discretized first, and then the differential algebra method is utilized to estimate the maneuvering parameter of the noncooperative target in the discretized time. Since the discretized period is very short, the maneuvering parameters of the target in the next discretized time are assumed to be the same as those in the previous discretized time, and the estimated maneuvering parameters are utilized to predict the target’s relative state in the next discretized time to achieve maneuvering trajectory tracking. Compared with the interactive multimodel method (IMM), the proposed method can estimate the maneuvering parameter of the noncooperative target in real time, which greatly reduces the tracking error caused by the mismatching of the target’s maneuvering model. In order to verify the effectiveness of the algorithm, a simulation of a noncooperative target’s maneuvering trajectory tracking is provided. The results demonstrated that the proposed method could track the noncooperative target maneuvering in real time, and the estimation accuracy was improved by about 93.07% compared with the IMM.

Interacting multiple model-based ETUKF for efficient state estimation of connected vehicles with V2V communication

Accurate prediction of the motion state of the connected vehicles, especially the preceding vehicle (PV), would effectively improve the decision-making and path planning of intelligent vehicles. The evolution of vehicle-to-vehicle (V2V) communication technology makes it possible to exchange data between vehicles. However, since V2V communication has a transmission interval, which will result in the host vehicle not receiving information from the PV within the time interval. Furthermore, V2V communication is a time-triggered system that may occupy more communication bandwidth than required. On the other hand, traditional estimation methods of the PV state based on individual models are usually not applicable to a wide range of driving conditions. To address these issues, an event-triggered unscented Kalman filter (ETUKF) is first employed to estimate the PV state to strike a balance between estimation accuracy and communication cost. Then, an interactive multi-model (IMM) approach is combined with ETUKF to form IMMETUKF to further improve the estimation accuracy and applicability. Finally, simulation experiments under different driving conditions are implemented to verify the effectiveness of IMMETUKF. The test results indicated that the IMMETUKF has high estimation accuracy even when the communication rate is reduced to 14.84% and the proposed algorithm is highly adaptable to different driving conditions.

Adaptive Markov IMM Based Multiple Fading Factors Strong Tracking CKF for Maneuvering Hypersonic-Target Tracking

Hypersonic targets have complex motion states and high maneuverability. The traditional interactive multi-model (IMM) has low tracking accuracy and a slow convergence speed. Therefore, this paper proposes a strong tracking cubature Kalman filter (CKF) adaptive interactive multi-model (AIMM) based on multiple fading factors. Firstly, this paper analyzes the structure of the CKF algorithm, introduces the fading factor of the strong tracking algorithm into the covariance matrix of the time update and measurement update, and adjusts the filter gain online and in real time, which can reduce the decline infilter accuracy caused by model mismatch. Secondly, Singer model, “current” statistical (CS) model, and Jerk model are selected in the model set of IMM and introduced singular value decomposition (SVD) decomposition to solve the problem that Cholesky decomposition cannot be performed in the CKF due to the model dimension expansion. Last, an adaptive algorithm for the Markov matrix in the IMM is proposed. The transition probability was adaptively modified by the value of the model likelihood function to enhance the proportion of matching models. The simulation results show that the proposed algorithm enhanced the proportion of matching models in the IMM and improved the tracking accuracy by 16.51% and the convergence speed by 37.5%.

Target Filter Tracking Algorithm Based on IABBSCA-IMM

Aiming at the problem that the probability transition matrix is set by prior information and the filtering parameters are fixed in the traditional interactive multimodel (IMM) algorithm, which leads to the model probability lag in the switching process and the low filtering and tracking accuracy, an interactive multimodel filter tracking (IABBSCA-IMM) algorithm with improved parameter adaptive and bare bone sine cosine optimization is proposed. First, the Markov probability transition matrix is dynamically adjusted and limited conditions are added through the parameter adaptation method; then, the filtering parameters Q and R are optimized by the bare bones sine cosine algorithm (BBSCA); finally, three motion models of CV (uniform velocity motion), CA (uniform acceleration motion), and CT (uniform velocity turning motion) are used to conduct filtering and tracking experiments on the target. The simulation results show that, compared with the IMM algorithm, the AMP-IMM algorithm, the IASCA-IMM algorithm, and the IABBSCA-IMM algorithm proposed in this study have the smallest position and velocity root mean square error (RMSE) in the X and Y directions, and the accuracy is better.

Adaptive Tracking of High-Maneuvering Targets Based on Multi-Feature Fusion Trajectory Clustering: LPI's Purpose.

Since the passive sensor has the property that it does not radiate signals, the use of passive sensors for target tracking is beneficial to improve the low probability of intercept (LPI) performance of the combat platform. However, for the high-maneuvering targets, its motion mode is unknown in advance, so the passive target tracking algorithm using a fixed motion model or interactive multi-model cannot match the actual motion mode of the maneuvering target. In order to solve the problem of low tracking accuracy caused by the unknown motion model of high-maneuvering targets, this paper firstly proposes a state transition matrix update-based extended Kalman filter (STMU-EKF) passive tracking algorithm. In this algorithm, the multi-feature fusion-based trajectory clustering is proposed to estimate the target state, and the state transition matrix is updated according to the estimated value of the motion model and the observation value of multi-station passive sensors. On this basis, considering that only using passive sensors for target tracking cannot often meet the requirements of high target tracking accuracy, this paper introduces active radar for indirect radiation and proposes a multi-sensor collaborative management model based on trajectory clustering. The model performs the optimal allocation of active radar and passive sensors by judging the accumulated errors of the eigenvalue of the error covariance matrix and makes the decision to update the state transition matrix according to the magnitude of the fluctuation parameter of the error difference between the prediction value and the observation value. The simulation results verify that the proposed multi-sensor collaborative target tracking algorithm can effectively improve the high-maneuvering target tracking accuracy to satisfy the radar’s LPI performance.

Stage of Charge Estimation of a Lithium-ion Battery Based on the Interactive Multi-model

Stage of Charge Estimation of a Lithium-ion Battery Based on the Interactive Multi-model

Interactive Multi-Model Fault Diagnosis Method of Switched Reluctance Motor Based on Low delay Anti-interference

Interactive Multi-Model Fault Diagnosis Method of Switched Reluctance Motor Based on Low delay Anti-interference

Variational Bayesian Based IMM Robust GPS Navigation Filter

This paper investigates the navigational performance of Global Positioning System (GPS) using the variational Bayesian (VB) based robust filter with interacting multiple model (IMM) adaptation as the navigation processor. The performance of the state estimation for GPS navigation processing using the family of Kalman filter (KF) may be degraded due to the fact that in practical situations the statistics of measurement noise might change. In the proposed algorithm, the adaptivity is achieved by estimating the time-varying noise covariance matrices based on VB learning using the probabilistic approach, where in each update step, both the system state and time-varying measurement noise were recognized as random variables to be estimated. The estimation is iterated recursively at each time to approximate the real joint posterior distribution of state using the VB learning. One of the two major classical adaptive Kalman filter (AKF) approaches that have been proposed for tuning the noise covariance matrices is the multiple model adaptive estimate (MMAE). The IMM algorithm uses two or more filters to process in parallel, where each filter corresponds to a different dynamic or measurement model. The robust Huber's M-estimation-based extended Kalman filter (HEKF) algorithm integrates both merits of the Huber M-estimation methodology and EKF. The robustness is enhanced by modifying the filter update based on Huber's M-estimation method in the filtering framework. The proposed algorithm, referred to as the interactive multi-model based variational Bayesian HEKF (IMM-VBHEKF), provides an effective way for effectively handling the errors with time-varying and outlying property of non-Gaussian interference errors, such as the multipath effect. Illustrative examples are given to demonstrate the navigation performance enhancement in terms of adaptivity and robustness at the expense of acceptable additional execution time.

Research on Channel Optimization of Ads-B Aviation Target Surveillance Radar Based on Improved Filtering Algorithm

In order to improve the surveillance effect of the aviation target surveillance radar, this paper improves the traditional filtering algorithm and builds the channel optimization system of the ADS-B aviation target surveillance radar based on the improved filtering algorithm. Moreover, this paper uses algorithm improvement to ensure the positive definite or semipositive definiteness of the state covariance and uses the root mean square volume Kalman filter to avoid the filter divergence or tracking interruption caused by the nonpositive definiteness of the matrix; the filtering principle of the interactive multimodel is to use multiple filters for parallel processing and achieve the adaptive adjustment algorithm residual error by adjusting the one-step prediction covariance in the adjustment algorithm. In addition, this paper combines the actual needs to construct a system functional structure to optimize the channel of the ADS-B aviation target surveillance radar and uses software engineering methods to model and analyze the requirements. Finally, this paper designs experiments to verify system performance. The research results show that the performance of the system constructed in this paper meets actual needs.

Using Camshift and Kalman Algorithm to Trajectory Characteristic Matching of Basketball Players

Because of its unique charm, sports video is widely welcomed by the public in today’s society. Therefore, the analysis and research of sports game video data have high practical significance and commercial value. Taking a basketball game video as an example, this paper studies the tracking feature matching of basketball players’ detection, recognition, and prediction in the game video. This paper is divided into four parts to improve the application of the interactive multimodel algorithm to track characteristic matching: moving object detection, recognition, basketball track characteristic matching, and player track characteristic matching. The main work and research results of each part are as follows: firstly, the improved K-means clustering algorithm is used to segment the golf field area; then, HSV is combined with the RGB Fujian value method to eliminate the field area; at last, straight field lines were extracted by Hough transform, and elliptical field lines were extracted by curve fitting, and the field lines were eliminated to realize the detection of moving objects. Seven normalized Hu invariant moments are used as the target features to realize the recognition of moving targets. By obtaining the feature distance between the sample and the template, the category of the sample is judged, which has a good robustness. The Kalman filter is used to match the characteristics of the basketball trajectory. Aiming at the occlusion of basketball, the least square method was used to fit the basketball trajectory, and the basketball position was predicted at the occlusion moment, which realized the occlusion trajectory matching. The matching of players’ track characteristics is realized by the CamShift algorithm based on the color model, which makes full use of players’ color information and realizes real-time performance. In order to solve the problem of occlusion between players in the track feature matching, CamShift and Kalman algorithms were used to determine the occlusion factor through the search window and then weighted Kalman and CamShift according to the occlusion degree to get the track feature matching result. The experimental results show that the detection time is greatly shortened, the memory space occupied is small, and the effect is very ideal.

Application of Improved Interactive Multimodel Algorithm in Player Trajectory Feature Matching

The original target tracking algorithm based on a single model has long been unable to meet the complex and changeable characteristics of the target, and then there are problems such as poor tracking accuracy, target loss, and model mismatch. The interactive multimodel algorithm uses multiple motion models to track the target, obtains the degree of adaptation between the actual motion state of the target and each model according to the calculated likelihood function, and then combines the updated weight values of each filter to obtain a weighted sum. Therefore, the interactive multimodel algorithm can achieve better performance. This paper proposes an improved interactive multimodel algorithm that can achieve player tracking and trajectory feature matching. First, this paper proposes an improved Kalman filtering (IKF) algorithm. This method is developed from the unbiased conversion measurement Kalman filter, which can obtain more accurate target state and covariance estimation. Secondly, using the parallel processing mode of the IMM algorithm to efficiently solve the data association between multiple filters, the IMM-IKF model is proposed. Finally, in order to solve the problem of low computational efficiency and high mismatch rate in image feature point matching, a method of introducing a minimum spanning tree in two-view matching is proposed. Experimental results show that the improved IMM-IKF algorithm can quickly respond to changes in the target state and can find the matching path with the lowest matching cost. In the case of ensuring the matching accuracy, the real-time performance of image matching is ensured.

Interactive Multi-Model Independent Joint Probabilistic Data Association Filtering Algorithm

Aiming at the problem of redundant calculation of the Interactive Multi-Model-Joint Probabilistic Data Association Filtering (IMM-JPDAF) algorithm, an Interactive Multi-Model Independent Joint Probabilistic Data Association Filtering algorithm is proposed. The algorithm uses model probabilities as priori information to reliably estimate the parameters of the maneuvering target motion model, and then uses them in the JPDA module, so that the JPDA can exist independently of the model filters. The simulation results show that the proposed algorithm can save about 20% of the calculation amount without reducing the accuracy of the target tracking, effectively reducing the redundancy.