In an intelligent transportation system (ITS), the single motion model is not enough to meet the needs of simulation systems (or practical applications) and guarantee good tracking performance for multiple maneuvering extended target tracking (ETT). This paper presents multiple models Poisson multi-Bernoulli mixture (MM-PMBM) filter for extended target tracking to address this problem. To make multiple models transform in an accurate manner, the jump Markov system (JMS) is timely introduced in filtering recursion. After that, the model probability of target is recursively updated and used to estimate the target position. In addition, we derive the closed-form solution to the proposed filter based on the Gamma Gaussian inverse-Wishart (GGIW) implementation. The approach to handling the merging problem of GGIW components via Kullback-Leibler divergence (KLD) minimization is also presented, resulting in better utilization of available components. Simulation results demonstrate that the proposed filter has superior performance in comparison to several state-of-the-art multi-target filters.

In a multi-target tracking (MTT) scenario, the prior knowledge of parameters such as clutter rate and detection probability are usually uncertain or estimated offline from training data. However, significant parameters mismatch in the clutter and detection model will result in biased estimates. In such cases, the ability to adaptively online estimate clutter rate and detection probability is critical in MTT under the random finite sets (RFS) framework. In this paper, we propose a robust Poisson multi-Bernoulli mixture (PMBM) filter that can accommodate model mismatch in clutter rate and detection probability for multiple extended target tracking problems. Moreover, the closed-form solution to the proposed method is derived by the use of Beta and Gamma Gaussian inverse-Wishart (GGIW) distribution, where Beta is used to describe unknown detection probability and GGIW is used to model extended target extension as an ellipse. Simulation results are presented to verify the effectiveness of the proposed method.

On the sea, scene parsing as one of the most basic environmental perception methods for autonomous tasks, helps autonomous devices safely navigate and avoid obstacles. However, most of the existing methods lack real-time performance and have difficulty to accurately detect small obstacles and changeable textures. We therefore propose a real-time scene parsing network for autonomous maritime transportation. Specifically, we first design a lightweight model framework, and then explore three efficient loss functions, such that a balance between accuracy and real-time performance can be achieved. The first loss function is an obstacle weighted loss, which improves the extraction of small obstacles by analyzing the distribution law of obstacles at sea. The second one is a detail loss, which optimizes the detail segmentation at complex contours by emphasizing edge features. The last one is an affinity loss, utilizing the context dependency between features to accurate detect reflections, ripples, and other changeable textures. In addition, a new maritime sense parsing dataset called GBA dataset is proposed and made publicly available. We tested the proposed model on GBA dataset and MaSTr1325 dataset, and the experimental results show that the proposed method achieves superior performance in both segmentation and speed, with mIOU of 94.59%, and FPS of 39.44.

This paper proposes a Poisson multi-Bernoulli mixture (PMBM) forward-backward smoother for robust multi-target tracking of point sources within the random finite sets (RFS) framework. The method operates in two phases: forward filtering and backward smoothing. During forward filtering, the PMBM density is recursively propagated via Bayesian prediction and update. This phase explicitly models target births, deaths, and existence uncertainties, even under high clutter densities. In the backward smoothing phase, state estimates are refined through parametric Bayesian retrodiction. Here, posterior PMBM weights are analytically updated using future measurements, effectively resolving temporal ambiguities. A simulation study is conducted to evaluate the performance of the proposed method, demonstrating its effectiveness in practical scenarios.

To address the high computational complexity and data association challenges in multi-sensor networks, this paper proposes a spatial-adaptive confidence-weighted Poisson multi-Bernoulli mixture (SACW-PMBM) algorithm that applies a divide-and-conquer fusion strategy, reformulating the global multi-target estimation problem into a series of localized, weighted tasks. Existing distributed PMBM fusion methods, such as those employing the weighted arithmetic average (WAA) or generalized covariance intersection (GCI), utilize a single scalar weight per sensor, which fails to capture spatially varying information quality. The primary contribution of this work is to replace these scalar weights with a state-dependent confidence function, ω^(s) (x), that dynamically models sensor reliability across the state space. This confidence function is constructed using a variational autoencoder (VAE) that jointly represents local information entropy and environmental disturbances. We subsequently derive a novel fusion criterion by minimizing the confidence-weighted Kullback-Leibler (KL) divergence between the local sensor posteriors and the fused global density. A computationally tractable hierarchical architecture is also developed to manage hypothesis complexity. Simulation results in challenging scenarios with severe occlusion and non-uniform clutter demonstrate that the proposed SACW-PMBM algorithm achieves superior tracking accuracy and robustness compared to benchmark centralized PMBM (C-PMBM), WAA, and GCI fusion methods.

This paper introduces an advanced extended object tracking algorithm developed within the random finite sets (RFS) framework, utilizing a Poisson multi-Bernoulli mixture (PMBM) filter with forward-backward smoothing capabilities. The proposed solution combines a forward PMBM filter propagation with a novel backward multi-Bernoulli (MB) smoother, establishing a complete Bayesian framework for extended object tracking. The methodology begins with a rigorous derivation of the standard PMBM forward recursion, followed by the development of an MB backward smoothing procedure. A key theoretical contribution is the demonstration that the backward smoothing density, obtained through set integration of survival object components, maintains the MB distribution property. For practical implementation, we derive a closed-form solution using Gamma Gaussian inverse-Wishart (GGIW) distributions, enabling efficient estimation of both kinematic states and spatial extents. Computational efficiency is further enhanced through strategic application of Murty’s algorithm for optimal hypothesis management. Comprehensive simulation studies validate the algorithm’s performance, demonstrating significant improvements in tracking accuracy and computational efficiency compared to conventional approaches. The effectiveness of the proposed method is validated via an extensive simulation study.