MCMC Particle Filter based Multiple Preceeding Vehicle Tracking System for Intelligent Vehicle

Baehoon Choi, Jhonghyun An, Minho Cho, Euntai KimJournal of Korean Institute of Intelligent Systems 2015 · DOI

Abstract

Precise environment recognition is a cornerstone of autonomous driving. In complex driving scenarios, vehicles must accurately track multiple dynamic obstacles, such as other cars and pedestrians, while overcoming sensor uncertainties. This paper proposes a multi-vehicle tracking algorithm using microwave radar. The system leverages Mixture Density Networks (MDN) to model the probabilistic distribution of radar measurements, specifically focusing on vehicle corners. To address the "curse of dimensionality" and manage a time-varying number of targets, the algorithm employs a Reversible Jump Markov Chain Monte Carlo (RJMCMC) sampling strategy within a particle filter framework. Simulation results demonstrate the robustness of the proposed approach.

Introduction

Tracking multiple objects (Multi-Target Tracking or MTT) is a fundamental challenge in intelligent vehicle systems. Radar sensors, while reliable in various weather conditions, produce noisy data that often clusters around specific vehicle geometries rather than the center of mass. Furthermore, the number of targets in a scene is dynamic, requiring a tracking framework that can adaptively adjust its state space dimension. Standard particle filters often struggle with these high-dimensional, variable-state problems.

Methodology

Radar Modeling via Mixture Density Networks (MDN)

Traditional tracking models often assume Gaussian noise centered on a point target. However, radar returns from vehicles are typically concentrated on the rear corners. To capture this multi-modal behavior, the researchers utilized Mixture Density Networks. The MDN outputs the parameters of a Gaussian mixture model (GMM) that represents the likelihood of radar measurements given a vehicle's pose. This allows the filter to prioritize returns from expected corner locations and effectively ignore environmental clutter.

RJMCMC Particle Filter

To manage the varying number of vehicles, the system integrates Reversible Jump MCMC into the particle filter's sampling step. Unlike fixed-dimension filters, RJMCMC allows the state vector to change size through "Birth" and "Death" moves.

The RJMCMC Particle Filter process, illustrating the transition between states and the integration of the RADAR-Vehicle model.

As shown in the flowchart above, the process includes:

  • Birth Moves: Proposing new vehicle states when unexplained radar detections appear.
  • Death Moves: Removing existing tracks that are no longer supported by incoming data.
  • Diffusion Moves: Updating the states of existing targets.
  • Accept/Reject Step: A Metropolis-Hastings-like criterion ensures that jumps between different dimensions (adding or removing cars) are statistically sound, maintaining the stability of the posterior distribution.

Key Results

The proposed algorithm was validated through computer simulations involving multiple vehicles in various lane configurations and intersection scenarios.

Simulation of the tracking system showing the radar field of view, detected points (red circles), and estimated vehicle positions (black rectangles).

The results highlighted several strengths:

  • Robust Multi-Target Tracking: The system successfully maintained tracks on multiple vehicles simultaneously, even as they entered and exited the sensor's field of view.
  • Clutter Rejection: By incorporating a Poisson-distributed noise model in simulations, the researchers demonstrated that the RJMCMC "Accept/Reject" mechanism effectively filters out false positives (ghost targets).
  • Adaptive State Estimation: The "Birth" and "Death" mechanisms allowed the filter to automatically adjust the number of tracked objects in real-time without manual intervention or hard-coded limits.

Conclusion

By combining the expressive power of Mixture Density Networks for sensor modeling with the dimensional flexibility of RJMCMC, this research provides a robust framework for multi-vehicle tracking. The system effectively handles the non-Gaussian nature of radar returns and the dynamic complexity of real-world traffic. Future work will focus on real-vehicle implementation and the integration of additional sensors like LiDAR and cameras to enhance environmental perception.

Summary updated

Abstract

지능형 자동차는 주변 환경에 대한 인식을 바탕으로 동작을 계획하고 움직인다. 따라서 정확한 환경 인식은 자율 주행 자동차의 필수 요소로 여겨진다. 차량의 주행 환경은 차량이나 보행자 같은 동적인 장애물이 다수 존재하여, 안전한 동작을 위해 이런 동적 장애물에 대한 인식이 정확하게 이루어져야 한다. 이를 위해 센서의 불확실성을 극복하는 일이 필수적이다. 본 논문에서는 레이더 센서를 이용하여 다수의 차량을 인식하고 추적하는 알고리즘을 제안한다. 제안된 추적 시스템은 몇 가지 특징을 갖는다. 레이더 센서가 차량을 계측할 때, 그 데이터가 양 모서리에서 주로 나타나는 특징을 혼합 밀도 네트워크로 표현하고, 이렇게 표현된 레이더 데이터의 확률적인 분포를 파티클 필터의 가중치 계산에 적용하여 추적 알고리즘을 수행하였다. 또한, 파티클 필터가 갖는 차원의 저주를 극복하고 시간의 흐름에 따라 그 숫자가 변화하는 다수 대상체의 상태를 예측하기 위해 가역 점프 마르코프 체인 몬테 카를로 (RJMCMC)를 통한 샘플링을 적용하였다. 제안된 알고리즘은 시뮬레이션을 통해 검증되었다. Intelligent vehicle plans motion and navigate itself based on the surrounding environment perception. Hence, the precise environment recognition is an essential part of self-driving vehicle. There exist many vulnerable road users (e.g. vehicle, pedestrians) on vehicular driving environment, the vehicle must percept all the dynamic obstacles accurately for safety. In this paper, we propose an multiple vehicle tracking algorithm using microwave radar. Our proposed system includes various special features. First, exceptional radar measurement model for vehicle, concentrated on the corner, is described by mixture density network (MDN), and applied to particle filter weighting. Also, to conquer the curse of dimensionality of particle filter and estimate the time-varying number of multi-target states, reversible jump markov chain monte carlo (RJMCMC) is used to sampling step of the proposed algorithm. The robustness of the proposed algorithm is demonstrated through several computer simulations.
BibTeX
@article{choi2015mcmc,
  author = {Baehoon Choi and Jhonghyun An and Minho Cho and Euntai Kim},
  title = {{MCMC Particle Filter based Multiple Preceeding Vehicle Tracking System for Intelligent Vehicle}},
  journal = {Journal of Korean Institute of Intelligent Systems},
  year = {2015},
  doi = {10.5391/jkiis.2015.25.2.186},
}
Figure