Abstract
Artificial Intelligence and Digital Twins play an integral role in driving innovation in the domain of intelligent driving. Long short-term memory (LSTM) is a leading driver in the field of lane change prediction for manoeuvre anticipation. However, the decision-making process of such models is complex and non-transparent, hence reducing the trustworthiness of the smart solution. This work presents an innovative approach and a technical implementation for explaining lane change predictions of layer normalized LSTMs using Layer-wise Relevance Propagation (LRP). The core implementation includes consuming live data from a digital twin on a German highway, live predictions and explanations of lane changes by extending LRP to layer normalized LSTMs, and an interface for communicating and explaining the predictions to a human user. We aim to demonstrate faithful, understandable, and adaptable explanations of lane change prediction to increase the adoption and trustworthiness of AI systems that involve humans. Our research also emphases that explainability and state-of-the-art performance of ML models for manoeuvre anticipation go hand in hand without negatively affecting predictive effectiveness.
This research was co-funded by the Bavarian Ministry of Economic Affairs, Regional Development and Energy, project Dependable AI, IBM Deutschland GmbH, and IBM Research, and was carried out within the Center for AI jointly founded by IBM and fortiss.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Step (2) in Fig. 1 is called the copy LRP rule. The copy LRP rule is a particular case of the LRP-\(\epsilon \) Rule, where one lower-level node and \(n\) upper-layer nodes exist, the weights are set to one, the bias is zero, and the activation function is linear.
References
Alber, M., et al.: Innvestigate neural networks! J. Mach. Learn. Res. 20(93), 1–8 (2019). http://jmlr.org/papers/v20/18-540.html
Krämmer, A., Christoph Schöller, D.G., Knoll, A.: Providentia - a large scale sensing system for the assistance of autonomous vehicles. In: Robotics: Science and Systems (RSS), Workshop on Scene and Situation Understanding for Autonomous Driving (2019). https://sites.google.com/view/uad2019/accepted-posters
Arras, L., et al.: Explaining and Interpreting LSTMs, pp. 211–238. International Publishing, Cham (2019). https://doi.org/10.1007/978-3-030-28954-6_11. https://doi.org/10.1007/978-3-030-28954-6_11
Arras, L., Montavon, G., Müller, K., Samek, W.: Explaining recurrent neural network predictions in sentiment analysis. CoRR (2017). http://arxiv.org/abs/1706.07206
Ba, J.L., Kiros, J.R., Hinton, G.E.: Layer normalization (2016)
Bach, S., Binder, A., Montavon, G., Klauschen, F., Müller, K.R., Samek, W.: On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLOS ONE 10(7), 1–46 (2015). https://doi.org/10.1371/journal.pone.0130140. https://doi.org/10.1371/journal.pone.0130140
Chen, Y., Dong, C., Palanisamy, P., Mudalige, P., Muelling, K., Dolan, J.M.: Attention-based hierarchical deep reinforcement learning for lane change behaviors in autonomous driving. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 1326–1334 (2019). https://doi.org/10.1109/CVPRW.2019.00172
Dang, H.Q., Fürnkranz, J., Biedermann, A., Hoepfl, M.: Time-to-lane-change prediction with deep learning. In: 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), pp. 1–7 (2017). https://doi.org/10.1109/ITSC.2017.8317674
El Marai, O., Taleb, T., Song, J.: Roads infrastructure digital twin: a step toward smarter cities realization. IEEE Network 35(2), 136–143 (2020)
Gallitz, O., De Candido, O., Botsch, M., Melz, R., Utschick, W.: Interpretable machine learning structure for an early prediction of lane changes. In: Farkaš, I., Masulli, P., Wermter, S. (eds.) ICANN 2020. LNCS, vol. 12396, pp. 337–349. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-61609-0_27
Gallitz, O., De Candido, O., Botsch, M., Utschick, W.: Interpretable feature generation using deep neural networks and its application to lane change detection. In: 2019 IEEE Intelligent Transportation Systems Conference (ITSC), pp. 3405–3411. IEEE (2019)
Guillemot, M., Heusele, C., Korichi, R., Schnebert, S., Chen, L.: Breaking batch normalization for better explainability of deep neural networks through layer-wise relevance propagation. CoRR (2020). https://arxiv.org/abs/2002.11018
Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput., 1735–80, December 1997. https://doi.org/10.1162/neco.1997.9.8.1735
Hui, L.Y.W., Binder, A.: BatchNorm decomposition for deep neural network interpretation. In: Rojas, I., Joya, G., Catala, A. (eds.) IWANN 2019. LNCS, vol. 11507, pp. 280–291. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-20518-8_24
Ioffe, S., Szegedy, C.: Batch normalization: Accelerating deep network training by reducing internal covariate shift. CoRR (2015). http://arxiv.org/abs/1502.03167
Khan, M.Q., Lee, S.: A comprehensive survey of driving monitoring and assistance systems. Sensors 19(11) (2019). https://doi.org/10.3390/s19112574. https://www.mdpi.com/1424-8220/19/11/2574
Kumar, S.A.P., Madhumathi, R., Chelliah, P.R., Tao, L., Wang, S.: A novel digital twin-centric approach for driver intention prediction and traffic congestion avoidance. J. Reliable Intell. Environ. 4(4), 199–209 (2018). https://doi.org/10.1007/s40860-018-0069-y
Patel, S., Griffin, B., Kusano, K., Corso, J.J.: Predicting future lane changes of other highway vehicles using RNN-based deep models. CoRR (2018). http://arxiv.org/abs/1801.04340v1
Schwalbe, G., Finzel, B.: Xai method properties: a (meta-)study. ArXiv abs/2105.07190 (2021)
Steyn, W.J., Broekman, A.: Development of a digital twin of a local road network: a case study. J. Testing Eval. 51(1) (2021)
Sundararajan, M., Taly, A., Yan, Q.: Axiomatic attribution for deep networks. CoRR (2017). http://arxiv.org/abs/1703.01365
Tang, J., Liu, F., Zhang, W., Ke, R., Zou, Y.: Lane-changes prediction based on adaptive fuzzy neural network. Expert Syst. Appl., 452–463 (2018). https://doi.org/10.1016/j.eswa.2017.09.025
Thevendran, H., Nagendran, A., Hydher, H., Bandara, A., Oruthota, U.: Deep learning and computer vision for IoT based intelligent driver assistant system. In: 2021 10th International Conference on Information and Automation for Sustainability (ICIAfS), pp. 340–345 (2021). https://doi.org/10.1109/ICIAfS52090.2021.9605823
Xing, Y., et al.: Driver lane change intention inference for intelligent vehicles: framework, survey, and challenges. IEEE Trans. Veh. Technol., 1, March 2019. https://doi.org/10.1109/TVT.2019.2903299
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Wehner, C., Powlesland, F., Altakrouri, B., Schmid, U. (2022). Explainable Online Lane Change Predictions on a Digital Twin with a Layer Normalized LSTM and Layer-wise Relevance Propagation. In: Fujita, H., Fournier-Viger, P., Ali, M., Wang, Y. (eds) Advances and Trends in Artificial Intelligence. Theory and Practices in Artificial Intelligence. IEA/AIE 2022. Lecture Notes in Computer Science(), vol 13343. Springer, Cham. https://doi.org/10.1007/978-3-031-08530-7_52
Download citation
DOI: https://doi.org/10.1007/978-3-031-08530-7_52
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-08529-1
Online ISBN: 978-3-031-08530-7
eBook Packages: Computer ScienceComputer Science (R0)