This paper presents a tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. It reviews optimal and suboptimal Bayesian algorithms for nonlinear/non-Gaussian tracking problems, with a focus on particle filters. Particle filters are sequential Monte Carlo methods based on point mass (or "particle") representations of probability densities, which can be applied to any state-space model and generalize traditional Kalman filtering methods. Several variants of the particle filter, such as SIR, ASIR, and RPF, are introduced within a generic framework of the sequential importance sampling (SIS) algorithm. These are discussed and compared with the standard EKF through an illustrative example. The paper begins with an introduction to the problem of tracking, which involves estimating the state of a system over time using a sequence of noisy measurements. The state-space approach is used to model dynamic systems, focusing on the state vector that contains all relevant information about the system. The measurement vector represents noisy observations related to the state vector. The state-space approach is convenient for handling multivariate data and nonlinear/non-Gaussian processes. The paper discusses optimal algorithms, including the Kalman filter and grid-based methods, which are suitable for linear Gaussian systems. However, these methods are intractable for nonlinear/non-Gaussian systems. The paper then presents suboptimal algorithms, including the extended Kalman filter (EKF), approximate grid-based methods, and particle filters. The EKF uses local linearization to approximate nonlinear systems, while particle filters use sequential importance sampling to approximate the posterior density. The paper also describes particle filtering methods, including the sequential importance sampling (SIS) algorithm, which forms the basis for most sequential Monte Carlo filters. The SIS algorithm represents the required posterior density function by a set of random samples with associated weights. The paper discusses the degeneracy problem in particle filters, where most particles have negligible weight, and presents methods to reduce this issue, such as resampling and choosing an appropriate importance density. The paper concludes with a discussion of other related particle filters, including the sampling importance resampling (SIR) filter, auxiliary sampling importance resampling (ASIR) filter, and regularized particle filter (RPF). These filters are derived from the SIS algorithm and are used for different applications in Bayesian tracking. The paper emphasizes the importance of choosing an appropriate importance density and resampling strategy to improve the performance of particle filters in nonlinear/non-Gaussian systems.This paper presents a tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. It reviews optimal and suboptimal Bayesian algorithms for nonlinear/non-Gaussian tracking problems, with a focus on particle filters. Particle filters are sequential Monte Carlo methods based on point mass (or "particle") representations of probability densities, which can be applied to any state-space model and generalize traditional Kalman filtering methods. Several variants of the particle filter, such as SIR, ASIR, and RPF, are introduced within a generic framework of the sequential importance sampling (SIS) algorithm. These are discussed and compared with the standard EKF through an illustrative example. The paper begins with an introduction to the problem of tracking, which involves estimating the state of a system over time using a sequence of noisy measurements. The state-space approach is used to model dynamic systems, focusing on the state vector that contains all relevant information about the system. The measurement vector represents noisy observations related to the state vector. The state-space approach is convenient for handling multivariate data and nonlinear/non-Gaussian processes. The paper discusses optimal algorithms, including the Kalman filter and grid-based methods, which are suitable for linear Gaussian systems. However, these methods are intractable for nonlinear/non-Gaussian systems. The paper then presents suboptimal algorithms, including the extended Kalman filter (EKF), approximate grid-based methods, and particle filters. The EKF uses local linearization to approximate nonlinear systems, while particle filters use sequential importance sampling to approximate the posterior density. The paper also describes particle filtering methods, including the sequential importance sampling (SIS) algorithm, which forms the basis for most sequential Monte Carlo filters. The SIS algorithm represents the required posterior density function by a set of random samples with associated weights. The paper discusses the degeneracy problem in particle filters, where most particles have negligible weight, and presents methods to reduce this issue, such as resampling and choosing an appropriate importance density. The paper concludes with a discussion of other related particle filters, including the sampling importance resampling (SIR) filter, auxiliary sampling importance resampling (ASIR) filter, and regularized particle filter (RPF). These filters are derived from the SIS algorithm and are used for different applications in Bayesian tracking. The paper emphasizes the importance of choosing an appropriate importance density and resampling strategy to improve the performance of particle filters in nonlinear/non-Gaussian systems.