For the purpose of track maintenance in a cluttered environment with some false alarms, false dismissals and measurement disturbances, this paper presents a new multi-object tracking procedure with uncertainty of the source of the measurement data returned from multi-sensor. The tracking method is a thrice deeply-fusion approach constructed by a primary fusion based on an improved probabilistic data association filter (IPDAF), a secondary fusion with historical motion trajectories, and a thrice fusion with road-markings. It incorporates the existence probabilities of the individual tracks and the variable number of objects based on the Bayesian estimation theory, which can improve the tracking performances effectively in an environment with high clutter density. A binary 2-D assignment is adopted for the optimal data association, which is established as a nonlinear optimization problem. In the motion modeling, it introduces multiple measurement models for different sensors into the method. Then, the estimation could be performed with greater reliability. The computational efficiency is satisfying and it can be used for real-time application, which is verified by two real test scenarios.
WangXXuLSunH, et al. On-road vehicle detection and tracking using MMW radar and monovision fusion. IEEE Trans Intell Transp Syst2016; 17(7): 2075–2084.
2.
ZhuBFerranteAKarlssonJ, et al. Fusion of sensors data in automotive radar systems: a Spectral Estimation Approach. arXiv preprint 2019; arXiv 1908.02504 2019.
3.
YiCZhangKPengN.A multi-sensor fusion and object tracking algorithm for self-driving vehicles. Proc IMechE, Part D: J. Automobile Engineering2019; 233(9): 2293–2300.
4.
KirubarajanTBar-ShalomYPattipatiKR.Multiassignment for tracking a large number of overlapping objects [and application to fibroblast cells]. IEEE Trans Aerosp Electron Syst2001; 37(1): 2–21.
5.
de FeoMGrazianoAMiglioliR, et al. IMMJPDA versus MHT and Kalman filter with NN correlation: performance comparison. IEE Proc Radar Sonar Navig1997; 144(2): 49–56.
6.
GengKZouWYinG, et al. Low-observable targets detection for autonomous vehicles based on dual-modal sensor fusion with deep learning approach. Proc IMechE, Part D: J Automobile Engineering2019; 233(9): 2270–2283.
7.
GarciaFMartinDde la EscaleraA, et al. Sensor fusion methodology for vehicle detection. IEEE Intell Transp Syst Mag2017; 9(1): 123–133.
8.
Chavez-GarciaROAycardO.Multiple sensor fusion and classification for moving object detection and tracking. IEEE Trans Intell Transp Syst2016; 17(2): 525–534.
9.
Bar-ShalomYLiXR.Estimation and tracking: Principles, techniques, and software. Norwood, MA: Artech House, 1993.
10.
Bar-ShalomYLiXR.Multitarget-multisensor tracking: principles and techniques. Storrs, CT: YBS Publishing, 1995.
11.
DeckerSCamusTEledathJ.Collision Sensing by Stereo Vision and radar Sensor Fusion. IEEE Trans Intell Transp Syst2009; 10(4): 606–614.
12.
Bar-ShalomYTseE.Tracking in a cluttered environment with probabilistic data association. Automatica1975; 11(5): 451–460.
13.
ZhouZWangJWangY, et al. Visual tracking using improved multiple instance learning with Co-training framework for moving robot. KSII Trans Internet Inf Syst2018; 12(11): 5496–5521.
14.
ZhouZGaoXXiaJ, et al. Multiple instance learning tracking based on Fisher linear discriminant with incorporated priors. Int J Adv Robot Syst2018; 15(1): 172988141775072.
15.
DimitrakopoulosGUdenLVarlamisI.The future of Intelligent Transport Systems: autonomous driving levels and enablers. Elsevier, 2020; pp.189–197.
16.
ZhouZZhangRZhuZ.Robust Kalman filtering with long short-term memory for image-based visual servo control. Multimed Tools Appl2019; 78(18): 26341–26371.
17.
FanJWangZLinM, et al. Research on vehicle attitude and heading reference system based on multisensor information fusion. Proc IMechE, Part D: J Automobile Engineering2020; 234(13): 3056–3067.
18.
NoackBSijsJReinhardtM, et al. Decentralized data fusion with inverse covariance intersection. Automatica2017; 79: 35–41.
19.
SteyerSLenkCKellnerD, et al. Grid-based object tracking with nonlinear dynamic state and shape estimation. IEEE Trans Intell Transp Syst2020; 21(7): 2874–2893.
20.
XueboJJiaBJiaolingZ, et al. Centralized fusion estimation for uncertain multisensor system based on LMI method. In: 2009 international conference on mechatronics and automation, 2009, pp.2383–2387.
21.
OdatEShammaJSClaudelC.Vehicle classification and speed estimation using combined passive infrared/ultrasonic sensors. IEEE Trans Intell Transp Syst2018; 19(5): 1593–1606.
22.
SingerRSeaR.New results in optimizing surveillance system tracking and data correlation performance in dense multitarget environments. IEEE Trans Automat Contr1973; 18(6): 571–582.
23.
PetrovskayaAThrunS.Model based vehicle detection and tracking for autonomous urban driving. Auton Robots2009; 26(2–3): 123–139.
24.
JoKLeeMSunwooM.Track Fusion and behavioral reasoning for moving vehicles based on curvilinear coordinates of roadway geometries. IEEE Trans Intell Transp Syst2018; 19(9): 3068–3075.
25.
Bar-ShalomY.Tracking methods in a multitarget environment. IEEE Trans Automat Contr1978; 23(4): 618–626.
26.
Bar-ShalomYDaumFHuangJ.The probabilistic data association filter. IEEE Control Syst Mag2009; 29(6): 82–100.
27.
Bar-ShalomYLiXRKirubarajanT.Estimation with applications to tracking and navigation: theory algorithms and software. Hoboken, New Jersey: John Wiley & Sons, 2004.
28.
MurphyKRussellS.Rao-Blackwellised particle filtering for dynamic Bayesian networks, on Sequential Monte Carlo methods in practice. New York, NY: Springer, 2001, pp.499–515.
29.
KreucherCKastellaKHeroAO.Multitarget tracking using the joint multitarget probability density. IEEE Trans Aerosp Electron Syst2005; 41(4): 1396–1414.
30.
YearyMBHavlicekJP.A new centralized sensor fusion-tracking methodology based on particle filtering for power-aware systems. IEEE Trans Instrum Meas2008; 57(10): 2377–2387.
31.
BaiLWangY.A sensor fusion framework using multiple particle filters for video-based navigation. IEEE Trans Intell Transp Syst2010; 11(2): 348–358.
32.
BoersYDriessenJNVerschureF, et al. A multi target track before detect application. IEEE 2003 Conference on Computer Vision and Pattern Recognition Workshop, 16–22 June 2003, pp.104–104, Madison, WI. New York: IEEE.
33.
VermaakJGodsillSJPerezP.Monte Carlo filtering for multi-target tracking and data association. IEEE Trans Aerosp Electron Syst2005; 41(1): 309–332.
34.
LaghmaraHLaurainTCudelC, et al. Heterogeneous sensor data fusion for multiple object association using belief functions. Inf Fusion2020; 57: 44–58.
35.
LiaoJXiangGCaoL, et al. The left-behind human detection and tracking system based on vision with multi-model fusion and microwave radar inside the bus. Proc IMechE, Part D: J. Automobile Engineering2020; 234(9): 2342–2354.
36.
ZhuZChenXZhangX, et al. Probabilistic Data Fusion for short-term traffic prediction with semiparametric density ratio model. IEEE Trans Aerosp Electron Syst2019; 20(7): 2459–2469.
37.
ScheelADietmayerK.Tracking multiple vehicles using a variational radar model. IEEE Trans Aerosp Electron Syst2019; 20(10): 3721–3736.
38.
PolychronopoulosATsogasMAmditisAJ, et al. Sensor Fusion for predicting vehicles’ path for Collision Avoidance Systems. IEEE Trans Aerosp Electron Syst2007; 8(3): 549–562.
39.
YeddanapudiMBar-ShalomYPattipatiK.IMM estimation for multitarget-multisensor air traffic surveillance. Proc IEEE1997; 85(1): 80–96.
40.
ChenJLeungHLoT, et al. A modified probabilistic data association filter in a real clutter environment. IEEE Trans Aerosp Electron Syst1996; 32(1): 300–313.
41.
AzizAM.A novel all-neighbor fuzzy association approach for multitarget tracking in a cluttered environment. Signal Process2011; 91(8): 2001–2015.
42.
MalakianK.Track deletion logic for maneuvering and nonmaneuvering targets. Signal and data processing of small targets 1993. Int Socr Optics Photon1993; 1954: 314–319.
43.
WangXMusickiD.Low elevation sea-surface target tracking using IPDA type filters. IEEE Trans Aerosp Electron Syst2007; 43(2): 759–774.
44.
BernardinKStiefelhagenR.Evaluating multiple object tracking performance: the CLEAR MOT metrics. J Image Video Proc2008; 2008: 1–10.
