Research Papers

Noise-Statistics Learning of Automotive-Grade Sensors Using Adaptive Marginalized Particle Filtering

[+] Author and Article Information
Karl Berntorp

Mitsubishi Electric Research Laboratories
Cambridge, MA 02139
e-mail: karl.o.berntorp@ieee.org

Stefano Di Cairano

Mitsubishi Electric Research Laboratories
Cambridge, MA 02139
e-mail: dicairano@ieee.org

Contributed by the Dynamic Systems Division of ASME for publication in the JOURNAL OF DYNAMIC SYSTEMS, MEASUREMENT,AND CONTROL. Manuscript received July 5, 2018; final manuscript received January 22, 2019; published online February 21, 2019. Assoc. Editor: Jongeun Choi.

J. Dyn. Sys., Meas., Control 141(6), 061009 (Feb 21, 2019) (10 pages) Paper No: DS-18-1317; doi: 10.1115/1.4042673 History: Received July 05, 2018; Revised January 22, 2019

This paper presents a method for real-time identification of sensor statistics especially aimed for low-cost automotive-grade sensors. Based on recent developments in adaptive particle filtering (PF) and under the assumption of Gaussian distributed noise, our method identifies the slowly time-varying sensor offsets and variances jointly with the vehicle state, and it extends to banked roads. While the method is primarily focused on learning the noise characteristics of the sensors, it also produces an estimate of the vehicle state. This can then be used in driver-assistance systems, either as a direct input to the control system or indirectly to aid other sensor-fusion methods. The paper contains verification against several simulation and experimental data sets. The results indicate that our method is capable of bias-free estimation of both the bias and the variance of each sensor, that the estimation results are consistent over different data sets, and that the computational load is feasible for implementation on computationally limited embedded hardware typical of automotive applications.

Copyright © 2019 by ASME
Your Session has timed out. Please sign back in to continue.


Gustafsson, F. , 2009, “ Automotive Safety Systems,” IEEE Signal Process. Mag., 26(4), pp. 32–47.
Reif, K. , and Dietsche, K.-H. , 2011, Bosch Automotive Handbook, Robert Bosch, Plochingen, Germany.
Johansen, T. A. , Petersen, I. , Kalkkuhl, J. , and Ludemann, J. , 2003, “ Gain-Scheduled Wheel Slip Control in Automotive Brake Systems,” IEEE Trans. Control Syst. Technol., 11(6), pp. 799–811.
Paden, B. , Cap, M. , Yong, S. Z. , Yershov, D. , and Frazzoli, E. , 2016, “ A Survey of Motion Planning and Control Techniques for Self-Driving Urban Vehicles,” IEEE Trans. Intell. Veh., 1(1), pp. 33–55.
Berntorp, K. , 2017, “ Path Planning and Integrated Collision Avoidance for Autonomous Vehicles,” American Control Conference (ACC), Seattle, WA, May 24–26.
Thrun, S. , Montemerlo, M. , Dahlkamp, H. , Stavens, D. , Aron, A. , Diebel, J. , Fong, P. , Gale, J. , Halpenny, M. , Hoffmann, G. , Lau, K. , Oakley, C. , Palatucci, M. , Pratt, V. , and Stang, P. , Strohband, S. , Dupont, C. , Jendrossek, L.-E. , Koelen, C. , Markey, C. , Rummel, C. , van Niekerk, J. , Jensen, E. , and Alessandrini, P. , Bradski, G. , Davies, B. , Ettinger, S. , Kaehler, A. , and Nefian, A. , and Mahoney, P. , 2006, “ Stanley: The Robot That Won the DARPA Grand Challenge,” J. Field Rob., 23(9), pp. 661–692.
Nilsson, J. , Falcone, P. , Ali, M. , and Sjöberg, J. , 2015, “ Receding Horizon Maneuver Generation for Automated Highway Driving,” Control Eng. Pract., 41, pp. 124–133.
Solyom, S. , and Hultén, J. , 2013, Physical Model-Based Yaw Rate and Steering Wheel Angle Offset Compensation, Springer, Berlin, pp. 51–58.
Gustafsson, F. , Ahlqvist, S. , Forssell, U. , and Persson, N. , 2001, “ Sensor Fusion for Accurate Computation of Yaw Rate and Absolute Velocity,” SAE Paper No. 2001-01-1064. http://citeseerx.ist.psu.edu/viewdoc/download?doi=
Berntorp, K. , 2016, “ Joint Wheel-Slip and Vehicle-Motion Estimation Based on Inertial, GPS, and Wheel-Speed Sensors,” IEEE Trans. Control Syst. Technol., 24(3), pp. 1020–1027.
Baffet, G. , Charara, A. , and Lechner, D. , 2009, “ Estimation of Vehicle Sideslip, Tire Force and Wheel Cornering Stiffness,” Control Eng. Pract., 17(11), pp. 1255–1264.
Berntorp, K. , and Di Cairano, S. , 2018, “ Tire-Stiffness and Vehicle-State Estimation Based on Noise-Adaptive Particle Filtering,” IEEE Trans. Control Syst. Technol. (in press).
Berntorp, K. , and Di Cairano, S. , 2018, “ Offset and Noise Estimation of Automotive-Grade Sensors Using Adaptive Particle Filtering,” American Control Conference (ACC), Milwaukee, WI, June 27–29.
Doucet, A. , and Johansen, A. M. , 2009, “ A Tutorial on Particle Filtering and Smoothing: Fifteen Years Later,” Handbook of Nonlinear Filtering, D. Crisan , and B. Rozovsky , eds., Oxford University Press, Oxford, UK.
Eidehall, A. , Schön, T. B. , and Gustafsson, F. , 2005, “ The Marginalized Particle Filter for Automotive Tracking Applications,” Intelligent Vehicles Symposium, Las Vegas, NV, June 6–8.
Lundquist, C. , Karlsson, R. , Özkan, E. , and Gustafsson, F. , 2014, “ Tire Radii Estimation Using a Marginalized Particle Filter,” IEEE Trans. Intell. Transp. Syst., 15(2), pp. 663–672.
Anderson, B. D. O. , and Moore, J. B. , 1979, Optimal Filtering, Prentice Hall, Englewood Cliffs, NJ.
Schön, T. B. , Gustafsson, F. , and Nordlund, P.-J. , 2005, “ Marginalized Particle Filters for Mixed Linear Nonlinear State-Space Models,” IEEE Trans. Signal Process., 53(7), pp. 2279–2289.
Bishop, C. M. , 2006, Pattern Recognition and Machine Learning, Springer-Verlag, New York.
Saha, S. , and Gustafsson, F. , 2012, “ Particle Filtering With Dependent Noise Processes,” IEEE Trans. Signal Process., 60(9), pp. 4497–4508.
Gillespie, T. , 1992, Fundamentals of Vehicle Dynamics, Society of Automotive Engineers, Warrendale, PA.
Rajamani, R. , 2006, Vehicle Dynamics and Control, Springer-Verlag, New York.
Ryu, J. , and Gerdes, J. C. , 2004, “ Estimation of Vehicle Roll and Road Bank Angle,” American Control Conference (ACC), Boston, MA, June 30–July 2.
Gustafsson, F. , 2010, Statistical Sensor Fusion, Utbildningshuset/Studentlitteratur, Lund, Sweden.
Murphy, K. P. , 2007, “ Conjugate Bayesian Analysis of the Gaussian Distribution,” UBC, Vancouver, BC, Canada, Report. https://www.cs.ubc.ca/~murphyk/Papers/bayesGauss.pdf
Özkan, E. , Šmídl, V. , Saha, S. , Lundquist, C. , and Gustafsson, F. , 2013, “ Marginalized Adaptive Particle Filtering for Nonlinear Models With Unknown Time-Varying Noise Parameters,” Automatica, 49(6), pp. 1566–1575.
Rao, C. R. , 2001, Linear Statistical Inference and Its Applications, Wiley, Hoboken, NJ.
Berntorp, K. , and Di Cairano, S. , 2017, “ Particle Gibbs With Ancestor Sampling for Identification of Tire-Friction Parameters,” IFAC World Congress, Toulouse, France, July 9–14.
Lindsten, F. , Jordan, M. I. , and Schön, T. B. , 2014, “ Particle Gibbs With Ancestor Sampling,” J. Mach. Learn. Res., 15(1), pp. 2145–2184. http://www.jmlr.org/papers/volume15/lindsten14a/lindsten14a.pdf


Grahic Jump Location
Fig. 1

A schematic of the single-track model and related notation

Grahic Jump Location
Fig. 3

Estimated standard deviations (black solid) and true values (gray dashed) of the lateral acceleration, yaw rate, and roll rate in simulation for N =100 particles

Grahic Jump Location
Fig. 4

The estimated bias (black solid) and true bias (gray dashed) in simulation for N =100 particles

Grahic Jump Location
Fig. 5

Estimated steering offset (black solid) and true offset (gray dashed) in simulation for N =100 particles. The results from 100 Monte Carlo trials are displayed, and the offset is 0.28 deg at the road side in all simulations.

Grahic Jump Location
Fig. 6

The estimated road bank and rate (black solid) and true values (gray dashed) in simulation for N =100 particles

Grahic Jump Location
Fig. 7

The estimated lateral velocity and the yaw rate (black solid) and true values (gray dashed) in simulation for N =100 particles. Both states are estimated accurately.

Grahic Jump Location
Fig. 8

The upper plot shows the estimated steering-wheel offset (black solid) for 100 Monte Carlo executions and the ground truth (gray dashed), as obtained by an offline optimization-based procedure, in experiments for N =500 particles. The lower plot displays a corresponding histogram of the error distribution accumulated over all time steps after the initial transients (25 s), for all 100 Monte Carlo runs. The average error is displayed in dashed vertical line.

Grahic Jump Location
Fig. 9

The estimated bias (upper) standard deviation (mid) for 100 Monte Carlo runs and the estimated yaw rate for N =500 particles in experiments. The measured yaw rate is the dash-dotted line and the true yaw rate as measured by the validation equipment is in dashed.

Grahic Jump Location
Fig. 10

Estimated (black solid) and measured (gray dashed) bank angle for one realization (upper), estimated bias for 100 Monte Carlo runs (mid), and the error distribution of the bank angle estimation over the 100 Monte Carlo runs (cf., Fig. 8) in experiments

Grahic Jump Location
Fig. 11

The estimated steering-wheel offset (black solid) for four different data sets and the ground truth (gray dashed), as obtained by an offline optimization-based procedure, in experiments

Grahic Jump Location
Fig. 12

The measured yaw rate (upper plot) and the error between the measured yaw rate and the estimated yaw rate with bias compensation for four data sets (cf., Fig. 11, in experiments for N =500 particles)

Grahic Jump Location
Fig. 13

Average computation time for one iteration of Algorithm 1 for varying number N of particles. The computation time is measured in matlab on a 2014 i5 2.8 GHz processor.



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In