IVAS: Facilitating Safe and Comfortable Driving with Intelligent Vehicle Audio Systems

Vehicle audio systems affect safety and comfortability of daily driving, as a driver’s mental or physical state can be significantly influenced by his or her audio perception during the maneuver. In this paper, we propose an intelligent vehicle audio systems (IVAS) by leveraging recent advances in automated driving, recommender system, and audio synthesis in an end-to-end fashion. Our contributions are multifold: (i) we design a deep deterministic policy gradients (DDPG) for adaptive control of an automated vehicle equipped with multi-sensors; (ii) we incorporate an unidirectional long short-term memory (LSTM) units of a recurrent neural network (RNN) with a recurrent output layer for audio synthesis; and (iii) we integrate above techniques into an end-to-end framework and pioneer the next generation vehicle audio system—IVAS. A psychological study is also performed to demonstrate the effectiveness of the proposed framework.

[1]  Fuchun Yang,et al.  Power flow analysis of compound planetary gear trains in flight leading edge flaps , 2015, ICM 2015.

[2]  Yuval Tassa,et al.  Continuous control with deep reinforcement learning , 2015, ICLR.

[3]  Gao Li,et al.  Numerical study of lancet furrows dynamics behaviour in greater horseshoe bats , 2016 .

[4]  Kamal Nasrollahi,et al.  Deep Pain: Exploiting Long Short-Term Memory Networks for Facial Expression Classification , 2017, IEEE Transactions on Cybernetics.

[5]  Andriy Temko,et al.  V2Hz: Music composition from wind turbine energy using a finite-state machine , 2017, 2017 28th Irish Signals and Systems Conference (ISSC).

[6]  Edilson de Aguiar,et al.  Facial expression recognition with Convolutional Neural Networks: Coping with few data and the training sample order , 2017, Pattern Recognit..

[7]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[8]  Srinivas Peeta,et al.  Cooperative Adaptive Cruise Control for Connected Autonomous Vehicles by Factoring Communication-Related Constraints , 2018, Transportation Research Part C: Emerging Technologies.

[9]  Jianxiong Feng,et al.  Design and power flow analysis for multi-speed automatic transmission with hybrid gear trains , 2016 .

[10]  Jian Wang,et al.  Cooperative Adaptive Cruise Control for a Platoon of Connected and Autonomous Vehicles considering Dynamic Information Flow Topology , 2018, Transportation Research Record: Journal of the Transportation Research Board.

[11]  Lei Lin,et al.  Opinion Mining at Scale: A Case Study of the First Self-driving Car Fatality , 2018, 2018 IEEE International Conference on Big Data (Big Data).