A Performance Evaluation of the Alpha-Beta (α-β) Filter Algorithm with Different Learning Models: DBN, DELM, and SVM

Abstract

In this paper, we present a new Multiple learning to prediction algorithm model model that used three different combinations of machine-learning methods to improve the accuracy of the α-β filter algorithm. The parameters of α and β were tuned in dynamic conditions instead of static conditions. The proposed system was designed to use the deep belief network (DBN), the deep extreme learning machine (DELM), and the SVM as three different learning algorithms. Then these learned parameters were trained by the machine-learning algorithms tuned to the α-β filter algorithm as a prediction module, and they gave the final predicted results. The MAE and RMSE were used to evaluate the performance of the proposed α-β filter with different learning algorithms. Each algorithm recorded different best-case accuracy results; for the DBN, we achieved 3.60 and 2.61; for the DELM, we obtained the best-case result of 3.90 and 2.81; and finally, for the SVM, 4.0 and 3.21 were attained in terms of the RMSE and MAE, respectively, as compared to 5.21 and 3.95. When assessed in comparison with the typical alpha–beta filter algorithm, the proposed system provided results with better accuracy.

A higher prediction accuracy–based alpha–beta filter algorithm using the feedforward artificial neural network

“…If we derive this algorithm, we must presume that a model with two‐interval states adequately estimates a system. First, we do initialisation, as is shown in Equations (14) and (15) [2, 16]. $$${m}_{k-1}={c}_{1}$$$ $$${n}_{k-1}={c}_{2}$$$ …”

“…In terms of the feed-forward-based learning of the α-β filter, in the beginning, we calculated the RMSE [16] for the reading of the sensor data and compared the output with the original sensor data, that is, temperature data and humidity data. The RMSE was calculated for the sensor reading, which was extremely high, that is, 5.22.…”

“…The main issue with the ELM is that it is not suitable for smaller datasets. Khan and Kim [16] proposed a novel method to improve the accuracy of the smoothing filter using DBN, SVM, and DELM. Jacobs et al [17] proposed a new algorithm called mixture-of-experts.…”

“…An α-β filter algorithm [16] is a prominent method, and is also the simplest method, for solving estimation problems, and smoothing problems and for control applications. It is derived from the KF algorithm [2].…”

Multiple Pedestrian Tracking in Dense Crowds Combined with Head Tracking

“…DeepSort [35] is a classic MOT algorithm that complies with the TDB paradigm, and it was improved from the SORT algorithm. In order to predict the trajectories of the targets, researchers usually use state estimation filters, like the Kalman filter [36] and Alpha-Beta filter [37,38]. Before running DeepSort, it is helpful to use an independent detector like YOLOV3 to detect the targets of interest in each frame of the video, then use the Kalman filter, Hungarian algorithm [39], feature extractor, and other components to comprehensively consider the motion law, appearance similarity, motion similarity, and other information of the target bounding boxes, with the step of Kalman prediction, matching, and Kalman updating to calculate the target motion tracklets.…”