Robust multi-target sensing/tracking in the Bayesian Occupancy Filter framework

نویسندگان

  • Kamel Mekhnacha
  • David Raulo
چکیده

We present the “Bayesian Occupancy Filter” (BOF) and the “Fast ClusteringTracking” algorithms as a framework for robust sensing and multi-target tracking using multiple sensors. Perceiving of the surrounding physical environment reliably is a major demanding in smart systems requiring a high level of safety such as car driving assistant, autonomous robots, and surveillance. The dynamic environment need to be perceived and modeled according to the sensor measurements which could be noisy. To fit such a requirement, we propose a hierarchical approach in which two filtering layers are used: (i) Robust grid-level sensor fusion using the “Bayesian Occupancy Filter” algorithm in order to construct an occupancy/velocity grid representation of the environment. (ii) Robust object-level tracking using the “Fast Clustering-Tracking”. RÉSUMÉ. Nous présentons les algorithmes “Bayesian Occupancy Filter” (BOF) et “Fast Clustering-Tracking” comme un cadre pour la perception robuste et le suivi de cibles multiples en utilisant plusieurs capteurs. Percevoir l’environnement physique d’une manière fiable est une exigence clef dans les systèmes intelligents nécessitant un grand degré de sécurité tels que l’assistante à la conduite automobile, la robotique autonome et la surveillance. L’environnement dynamique doit être perçu et modélisé en utilisant les mesures capteurs généralement bruitées. Pour répondre à cette exigence, nous proposons une approche hiérarchique dans laquelle deux niveaux de filtrage sont utilisés: (i) Fusion robuste de capteurs en utilisant l’algorithme “Bayesian Occupancy Filter” dans lequel l’environnement dynamique est représenté par une grille d’occupation/vitesse. (ii) Suivi robuste d’objets en utilisant l’algorithme “Fast Clustering-Tracking”.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Bayesian Programming for Multi-target Tracking: An Automotive Application

gramme " La Route Automatisée " (http://www.lara.prd.fr/) and the Euro-pean project IST-1999-12224 " Sensing of Car Environment at Low Speed Driving " Abstract— A prerequisite to the design of future Advanced Driver Assistance Systems for cars is a sensing system providing all the information required for high-level driving assistance tasks. In particular, target tracking is still challenging i...

متن کامل

Clutter Removal in Sonar Image Target Tracking Using PHD Filter

In this paper we have presented a new procedure for sonar image target tracking using PHD filter besides K-means algorithm in high density clutter environment. We have presented K-means as data clustering technique in this paper to estimate the location of targets. Sonar images target tracking is a very good sample of high clutter environment. As can be seen, PHD filter because of its special f...

متن کامل

Estimation of LOS Rates for Target Tracking Problems using EKF and UKF Algorithms- a Comparative Study

One of the most important problem in target tracking is Line Of Sight (LOS) rate estimation for using from PN (proportional navigation) guidance law. This paper deals on estimation of position and LOS rates of target with respect to the pursuer from available noisy RF seeker and tracker measurements. Due to many important for exact estimation on tracking problems must target position and Line O...

متن کامل

The Bayesian occupation filter

Perception of and reasoning about dynamic environments is pertinent for mobile robotics and still constitutes one of the major challenges. To work in these environments, the mobile robot must perceive the environment with sensors; measurements are uncertain and normally treated within the estimation framework. Such an approach enables the mobile robot to model the dynamic environment and follow...

متن کامل

Study on Multi-Target Tracking Based on Particle Filter Algorithm

Particle filter is a probability estimation method based on Bayesian framework and it has unique advantage to describe the target tracking non-linear and non-Gaussian. In this study, firstly, analyses the particle degeneracy and sample impoverishment in particle filter multi-target tracking algorithm and secondly, it applies Markov Chain Monte Carlo (MCMC) method to improve re-sampling process ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2008