Cheap Joint Probabilistic Data Association with Adaptive Neuro-Fuzzy Inference System State Filter for Tracking Multiple Targets in Cluttered Environment

Fuzzy region assignment for visual tracking

“…There are systems that have available plenty of both data input and outputs, as there is a historical data scenario. In this situation, using automatic learning techniques, the system can be modeled and adjusted, in a similar way as the works based on neural networks (Zhu 1994) or neurofuzzy systems (Turkmen et al 2004) to approximate the assignment probabilities. There are other situations in which these data are not available or are very partial.…”

“…Thus, non-algorithmic approaches such as artificial neural networks, fuzzy systems and genetic algorithms can be applied to data association problems, isolated or in conjunction with classical formulations. Methods based on fuzzy systems and artificial n eural n etworks h ave been used to compute the association probabilities in JPDAF, to take the best decisions in the association process in different conditions, accordingly to the characteristics of objects and available sensors (Turkmen et al 2004;Sengupta et al 1989;Chen et al 2001). Genetic Algorithms, with a recognized capability to address hard search problems, have been previously applied in the data association problem in radar data processing by Angus et al (1993) and by Hillis (1997) to deal with the mono and multiscan data association problems, respectively.…”

“…Neural networks have been applied to estimate the association probabilities in JPDA filters (Sengupta et al 1989;Turkmen et al 2004;Zhu et al 1994;Shams 1996), representing the measured residuals between observation and tracks as inputs for the network. These approaches have proved capable of handling complex scenes with radar data, although the implementations require a large number of neurons and the preparation of large training data sets to have a reliable system.…”

A Hybrid Multi-sensor Multi-target Tracking Scheme with MLE and ANFIS

“…Therefore, for multi-sensor multi-target measurements, a hybrid scheme is proposed, which firstly partition of the same target, then applying for a single sensor multi-target tracking. JPDA algorithm [10][11][12][13] was proposed, which is based on adaptive neuro-fuzzy inference system and cheap JPDA algorithm, however, this method is only suitable for a single sensor. In this paper, a hybrid multiple sensor multiple target tracking system is presented, which applied maximum likelihood estimation, adaptively neuro-fuzzy inference system [14][15][16], and cheap JPDA algorithm.…”

Resonant Frequency Calculation for Circular Microstrip Antennas With a Dielectric Cover Using Adaptive Network-Based Fuzzy Inference System Optimized by Various Algorithms

“…In previous works [29][30][31][32][33][34][35][36][37][38], we successfully utilized ANFIS for computing accurately the various parameters of the rectangular, triangular, and circular MSAs, and for tracking multiple targets and estimating the phase inductance of the switched reluctance motors. In reference [32], the resonant frequency of circular MSA without a dielectric cover has been computed by using ANFIS.…”