access icon free Practical classification of different moving targets using automotive radar and deep neural networks

© The Institution of Engineering and Technology
Received 08/03/2018, Accepted 05/04/2018, Published 06/04/2018

In this work, the authors present results for classification of different classes of targets (car, single and multiple people, bicycle) using automotive radar data and different neural networks. A fast implementation of radar algorithms for detection, tracking, and micro-Doppler extraction is proposed in conjunction with the automotive radar transceiver TEF810X and microcontroller unit SR32R274 manufactured by NXP Semiconductors. Three different types of neural networks are considered, namely a classic convolutional network, a residual network, and a combination of convolutional and recurrent network, for different classification problems across the four classes of targets recorded. Considerable accuracy (close to 100% in some cases) and low latency of the radar pre-processing prior to classification (∼0.55 s to produce a 0.5 s long spectrogram) are demonstrated in this study, and possible shortcomings and outstanding issues are discussed.

Inspec keywords: recurrent neural nets; radar transmitters; road vehicle radar; electrical engineering computing; radar receivers; radar tracking; object tracking; Doppler radar; object detection; signal classification; microcontrollers; radar detection

Other keywords: radar detection; microDoppler extraction; radar pre-processing; deep neural network; NXP Semiconductors; automotive radar transceiver TEF810X; radar tracking; microcontroller unit SR32R274; recurrent neural network; target classification; residual neural network; convolutional neural network; time 0.5 s

Subjects: Microprocessor chips; Microprocessors and microcomputers; Neural computing techniques; Digital signal processing; Radar equipment, systems and applications; Signal detection; Electrical engineering computing

References

    1. 1)
      • 15. Gashinova, M., Hoare, E., Stove, A.: ‘Predicted sensitivity of a 300 GHz FMCW radar to pedestrians’. 2016 European Radar Conf. (EuRAD), London, 2016, pp. 350–353.
    2. 2)
      • 16. Belgiovane, D., Chen, C.C.: ‘Micro-Doppler characteristics of pedestrians and bicycles for automotive radar sensors at 77 GHz’. 2017 11th European Conf. on Antennas and Propagation (EUCAP), Paris, 2017, pp. 2912–2916.
    3. 3)
      • 32. Khashbat, J., Tsevegjav, T., Myagmarjav, J., et al: ‘Determining the driver's reaction time in the stationary and real-life environments (comparative study)’. 2012 7th Int. Forum on Strategic Technology (IFOST), 2012.
    4. 4)
      • 23. Jokanovic, B., Amin, M.G., Ahmad, F.: ‘Effect of data representations on deep learning in fall detection’. 2016 IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM), Rio de Janerio, 2016, pp. 1–5.
    5. 5)
      • 12. Lee, S., Yoon, Y.J., Lee, J.E., et al: ‘Human–vehicle classification using feature-based SVM in 77-GHz automotive FMCW radar’, IET Radar Sonar Navig., 2017, 11, (10), pp. 1589–1596.
    6. 6)
      • 35. Srivastava, N., Hinton, G., Krizhevsky, A., et al: ‘Dropout: a simple way to prevent neural networks from overfitting’. J. Mach. Learn. Res., 2014, 15, pp. 1929–1958.
    7. 7)
      • 26. Parashar, K.N., Oveneke, M.C., Rykunov, M., et al: ‘Micro-Doppler feature extraction using convolutional auto-encoders for low latency target classification’. 2017 IEEE Radar Conf. (RadarConf), Seattle, WA, 2017, pp. 1739–1744.
    8. 8)
      • 29. Klarenbeek, G., Harmanny, R.I.A., Cifola, L.: ‘Multi-target human gait classification using LSTM recurrent neural networks applied to micro-Doppler’. 2017 European Radar Conf. (EURAD), Nuremberg, 2017, pp. 167–170.
    9. 9)
      • 8. Saebboe, J., Viikari, V., Varpula, T., et al: ‘Harmonic automotive radar for VRU classification’. 2009 Int. Radar Conf. on ‘Surveillance for a Safer World’ (RADAR 2009), Bordeaux, 2009, pp. 1–5.
    10. 10)
      • 5. Hasch, J.: ‘Driving towards 2020: automotive radar technology trends’. 2015 IEEE MTT-S Int. Conf. on Microwaves for Intelligent Mobility (ICMIM), Heidelberg, 2015, pp. 1–4.
    11. 11)
      • 18. Stolz, M., Schubert, E., Meinl, F., et al: ‘Multi-target reflection point model of cyclists for automotive radar’. 2017 European Radar Conf. (EURAD), Nuremberg, 2017, pp. 94–97.
    12. 12)
      • 11. Lee, J., Kim, D., Jeong, S., et al: ‘Target classification scheme using phase characteristics for automotive FMCW radar’. IET Electron. Lett., 2016, 52, (25), pp. 2061–2063.
    13. 13)
      • 2. SAE International: ‘Taxonomy and definitions for terms related to driving automation systems for On-road motor vehicles’. Available athttp://standards.sae.org/j3016_201609/, 2016, accessed February 2018.
    14. 14)
      • 25. Seyfioğlu, M.S., Gürbüz, S.Z., Özbayoğlu, A.M., et al: ‘Deep learning of micro-Doppler features for aided and unaided gait recognition’. 2017 IEEE Radar Conf. (RadarConf), Seattle, WA, 2017, pp. 1125–1130.
    15. 15)
      • 10. Schubert, E., Meinl, F., Kunert, M., et al: ‘Clustering of high-resolution automotive radar detections and subsequent feature extraction for classification of road users’. 2015 16th Int. Radar Symp. (IRS), 2015, pp. 174–179.
    16. 16)
      • 19. Tahmoush, D.: ‘Review of micro-Doppler signatures’, IET Radar Sonar Navig., 2015, 9, (9), pp. 1140–1146.
    17. 17)
      • 28. Trommel, R.P., Harmanny, R.I.A., Cifola, L., et al: ‘Multi-target human gait classification using deep convolutional neural networks on micro-Doppler spectrograms’. Radar Conf. (EuRAD), 2016 European, 2016, pp. 81–84.
    18. 18)
      • 34. Munkres, J.: ‘Algorithms for the assignment and transportation problems’, J. Soc. Ind. Appl. Math., 1957, 5, (1), pp. 32–38.
    19. 19)
      • 24. Jokanovic, B., Amin, M., Erol, B.: ‘Multiple joint-variable domains recognition of human motion’. 2017 IEEE Radar Conf. (RadarConf), Seattle, WA, 2017, pp. 0948–0952.
    20. 20)
      • 22. Jokanović, B., Amin, M.: ‘Fall detection using deep learning in range-Doppler radars’, IEEE Trans. Aerosp. Electron. Syst., PP, (99), 2018, 52, (1), pp. 180–189.
    21. 21)
      • 30. Patel, J.S., Fioranelli, F., Ritchie, M., et al: ‘Multistatic radar classification of armed vs unarmed personnel using neural networks’, Evol. Syst., 2017, pp. 1–10.
    22. 22)
      • 3. ‘Expect the unexpected – an IET transport sector report on the unintended consequences of connected and autonomous vehicles’ (Institution of Engineering and Technology, 2017) accessed February 2018.
    23. 23)
      • 6. Heuel, S., Rohling, H.: ‘Two-stage pedestrian classification in automotive radar systems’. 2011 12th Int. Radar Symp. (IRS), 2011, pp. 477–484.
    24. 24)
      • 37. Ioffe, S., Szegedy, C.: ‘Batch normalization: accelerating deep network training by reducing internal covariate shift’, arXiv:1502.03167v3.
    25. 25)
      • 27. Seyfioğlu, M.S., Gürbüz, S.Z.: ‘Deep neural network initialization methods for micro-Doppler classification with low training sample support’, IEEE Geosci. Remote Sens. Lett., 2017, 14, (12), pp. 2462–2466.
    26. 26)
      • 7. Heuel, S., Rohling, H.: ‘Two-stage pedestrian classification in automotive radar systems’. 2012 13th Int. Radar Symp. (IRS), 2012, pp. 39–44.
    27. 27)
      • 13. Marchetti, E., Du, R., Noruzian, F., et al: ‘Radar reflectivity and motion characteristics of pedestrians at 300 GHz’. 2017 European Radar Conf. (EURAD), Nuremberg, 2017, pp. 57–60.
    28. 28)
      • 21. Kim, Y., Moon, T.: ‘Human detection and activity classification based on micro-Doppler signatures using deep convolutional neural networks’, IEEE Geosci. Remote Sens. Lett., 2016, 13, (1), pp. 8–12.
    29. 29)
      • 20. Fioranelli, F., Ritchie, M., Gürbüz, S., et al: ‘Feature diversity for optimized human micro-Doppler classification using multistatic radar’, IEEE Trans. Aerosp. Electron. Syst., 2017, 53, (2), pp. 640–654.
    30. 30)
      • 33. Rohling, H.: ‘Radar CFAR thresholding in clutter and multiple target situations’, IEEE Trans. Aerosp. Electron. Syst., 1983, AES-19, (4), pp. 608–621.
    31. 31)
      • 9. Sorowka, P., Rohling, H.: ‘Pedestrian classification with 24 GHz chirp sequence radar’. 2015 16th Int. Radar Symp. (IRS), 2015, pp. 167–173.
    32. 32)
      • 36. Hochreiter, S., Schmidhuber, J.: ‘Long short-term memory’, Neural Comput., 1997, 9, pp. 1735–1780.
    33. 33)
      • 31. He, K., Zhang, X., Ren, S., et al: ‘Deep residual learning for image recognition’, arXiv:1512.03385, 2015.
    34. 34)
      • 17. Belgiovane, D., Chen, C.C.: ‘Bicycles and human riders backscattering at 77 GHz for automotive radar’. 2016 10th European Conf. on Antennas and Propagation (EuCAP), Davos, 2016, pp. 1–5.
    35. 35)
      • 1. Jenn, U.: ‘The road to driverless cars: 1925–2025’. Available athttp://www.engineering.com/DesignerEdge/DesignerEdgeArticles/ArticleID/12665/The-Road-to-Driverless-Cars-1925-2025.aspx, 2016, accessed February 2018.
    36. 36)
      • 4. ‘Radar, camera, lidar, and V2X for autonomous cars’. Available athttps://blog.nxp.com/automotive/radar-camera-and-lidar-for-autonomous-cars, 2017, accessed February 2018.
    37. 37)
      • 14. Marchetti, E., Du, R., Noruzian, F., et al: ‘Comparison of pedestrian reflectivities at 24 and 300 GHz’. 2017 18th Int. Radar Symp. (IRS), Prague, 2017, pp. 1–7.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-rsn.2018.0103
Loading

Related content

content/journals/10.1049/iet-rsn.2018.0103
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading