Deep-Reinforcement-Learning-Based Distributed Vehicle Position Controls for Coverage Expansion in mmWave V2X

In millimeter wave (mmWave) vehicular communications, multi-hop relay disconnection by line-of-sight (LOS) blockage is a critical problem, especially in the early diffusion phase of mmWave-available vehicles, where not all the vehicles have mmWave communication devices. This paper proposes a distributed position control method for autonomous vehicles to make long relays connecting to road side units (RSUs) by avoiding blockages to communicate with each other via LOS paths. Even though vehicles with the proposed method do not use the whole information of the environments and cooperate with each other, they can decide their action (e.g., lane change and overtaking) to form long relays using only information of its surroundings (e.g., surrounding vehicle positions). The decision-making problem is formulated as a Markov decision process so that autonomous vehicles can learn a practical movement strategy of making long relays by a reinforcement learning (RL) algorithm. This paper designs a learning algorithm based on a sophisticated deep reinforcement learning algorithm, asynchronous advantage actor-critic (A3C), which enables vehicles to learn a complex movement strategy quickly by its deepneural-network architecture and multi-agent-learning mechanism. Once the strategy is well trained, vehicles can distributedly move to positions where the long relay to the RSU is established. Simulations results confirm that the proposed method can increase the relay length and coverage even if the traffic conditions and penetration ratio of mmWave communication devices in learning and operation phases are different.

[1]  Richard M. Murray,et al.  Information flow and cooperative control of vehicle formations , 2004, IEEE Transactions on Automatic Control.

[2]  V. Semkin,et al.  E-band propagation channel measurements in an urban street canyon , 2015, 2015 9th European Conference on Antennas and Propagation (EuCAP).

[3]  P. Raja,et al.  Optimal path planning of mobile robots: A review , 2012 .

[4]  Shane Legg,et al.  Massively Parallel Methods for Deep Reinforcement Learning , 2015, ArXiv.

[5]  Andreas Meier,et al.  Design of 5.9 ghz dsrc-based vehicular safety communication , 2006, IEEE Wireless Communications.

[6]  Kei Sakaguchi,et al.  Millimeter wave beamforming based on WiFi fingerprinting in indoor environment , 2015, 2015 IEEE International Conference on Communication Workshop (ICCW).

[7]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[8]  Tarik Taleb,et al.  Dynamic Clustering-Based Adaptive Mobile Gateway Management in Integrated VANET — 3G Heterogeneous Wireless Networks , 2011, IEEE Journal on Selected Areas in Communications.

[9]  Ch. Ramesh Babu,et al.  Internet of Vehicles: From Intelligent Grid to Autonomous Cars and Vehicular Clouds , 2016 .

[10]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[11]  Javier Gozalvez,et al.  LTE-V for Sidelink 5G V2X Vehicular Communications: A New 5G Technology for Short-Range Vehicle-to-Everything Communications , 2017, IEEE Vehicular Technology Magazine.

[12]  Robert W. Heath,et al.  Analysis of Urban Millimeter Wave Microcellular Networks , 2016, 2016 IEEE 84th Vehicular Technology Conference (VTC-Fall).

[13]  Koichi Ogawa,et al.  Path-Loss Prediction Models for Intervehicle Communication at 60 GHz , 2008, IEEE Transactions on Vehicular Technology.

[14]  Xiangyu Yu,et al.  A Novel Virtual Force Approach for Node Deployment in Wireless Sensor Network , 2012, 2012 IEEE 8th International Conference on Distributed Computing in Sensor Systems.

[15]  Prithwish Basu,et al.  Movement control algorithms for realization of fault-tolerant ad hoc robot networks , 2004, IEEE Network.

[16]  Alex Graves,et al.  Asynchronous Methods for Deep Reinforcement Learning , 2016, ICML.

[17]  Masahiro Morikura,et al.  Machine-Learning-Based Throughput Estimation Using Images for mmWave Communications , 2017, 2017 IEEE 85th Vehicular Technology Conference (VTC Spring).

[18]  Andrea Baiocchi,et al.  Road Side Unit coverage extension for data dissemination in VANETs , 2012, 2012 9th Annual Conference on Wireless On-Demand Network Systems and Services (WONS).

[19]  Xuemin Shen,et al.  Connected Vehicles: Solutions and Challenges , 2014, IEEE Internet of Things Journal.

[20]  Robert W. Heath,et al.  Radar aided beam alignment in MmWave V2I communications supporting antenna diversity , 2016, 2016 Information Theory and Applications Workshop (ITA).

[21]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[22]  Mario Gerla,et al.  Vehicular Cloud Computing , 2012, 2012 The 11th Annual Mediterranean Ad Hoc Networking Workshop (Med-Hoc-Net).

[23]  Saeid Nahavandi,et al.  System Design Perspective for Human-Level Agents Using Deep Reinforcement Learning: A Survey , 2017, IEEE Access.

[24]  Akihito Taya,et al.  Coverage Expansion through Dynamic Relay Vehicle Deployment in mmWave V2I Communications , 2018, 2018 IEEE 87th Vehicular Technology Conference (VTC Spring).

[25]  Jan Tommy Gravdahl,et al.  Modelling of UAV formation flight using 3D potential field , 2008, Simul. Model. Pract. Theory.

[26]  Kyungwhoon Cheun,et al.  Millimeter-wave beamforming as an enabling technology for 5G cellular communications: theoretical feasibility and prototype results , 2014, IEEE Communications Magazine.

[27]  Urbano Nunes,et al.  Platooning With IVC-Enabled Autonomous Vehicles: Strategies to Mitigate Communication Delays, Improve Safety and Traffic Flow , 2012, IEEE Transactions on Intelligent Transportation Systems.

[28]  Robert W. Heath,et al.  Millimeter Wave Vehicular Communications: A Survey , 2016, Found. Trends Netw..

[29]  Yusheng Ji,et al.  Cooperative Content Delivery in Vehicular Networks with Integration of Sub-6 GHz and mmWave , 2017, 2017 IEEE Globecom Workshops (GC Wkshps).

[30]  Roberto Verdone,et al.  Multihop R-ALOHA for intervehicle communications at millimeter waves , 1997 .

[31]  Richard M. Murray,et al.  DISTRIBUTED COOPERATIVE CONTROL OF MULTIPLE VEHICLE FORMATIONS USING STRUCTURAL POTENTIAL FUNCTIONS , 2002 .

[32]  Stephen Gundry,et al.  Autonomous and fault tolerant vehicular self deployment mechanisms in MANETs , 2013, 2013 IEEE International Conference on Technologies for Homeland Security (HST).

[33]  Ying Li,et al.  ChainCluster: Engineering a Cooperative Content Distribution Framework for Highway Vehicular Communications , 2014, IEEE Transactions on Intelligent Transportation Systems.

[34]  Hui Deng,et al.  Platoon management with cooperative adaptive cruise control enabled by VANET , 2015, Veh. Commun..

[35]  Jiming Chen,et al.  Novel Deployment Schemes for Mobile Sensor Networks , 2007, Sensors.

[36]  Rui Jiang,et al.  The adaptive cruise control vehicles in the cellular automata model , 2006 .