Interconnected Intelligent cars Sachin Sharma University of Hamburg - PowerPoint PPT Presentation

MIN Faculty Department of Informatics Interconnected Intelligent cars Sachin Sharma University of Hamburg Faculty of Mathematics, Informatics and Natural Sciences Department of Informatics Technical Aspects of Multimodal Systems 18. December 2017 Sachin Sharma – Interconnected Intelligent cars 1 / 30

Outline Introduction Motivation Approach Features Applications Challenges Conclusion 1. Introduction 2. Motivation 3. Approach 4. Features 5. Applications 6. Challenges 7. Conclusion Sachin Sharma – Interconnected Intelligent cars 2 / 30

Introduction Introduction Motivation Approach Features Applications Challenges Conclusion What does interconnected Intelligent cars means? ◮ Autonomous vehicles ◮ Inter car communication ◮ Mostly uses dedicated short range communication(DSRC) ◮ IoT(Internet of things) Sachin Sharma – Interconnected Intelligent cars 3 / 30

Motivation Introduction Motivation Approach Features Applications Challenges Conclusion ◮ Intelligent transport System(ITS) or Smart city ◮ Around 1.25 million deaths due to car accidents worldwide (WHO 2015 1 ) ◮ Up to 50 million nonfatal injuries each year as a result of road traffic crashes (WHO 2015 2 ) ◮ An increase in average speed of 1 km/h typically results in a 3% higher risk of a crash involving injury.(WHO 2004 3 ) 1 http://apps.who.int/iris/bitstream/10665/189242/1/9789241565066_eng.pdf?ua=1 2 http://apps.who.int/iris/bitstream/10665/189242/1/9789241565066_eng.pdf?ua=1 3 http://apps.who.int/iris/bitstream/10665/42871/1/9241562609.pdf Sachin Sharma – Interconnected Intelligent cars 4 / 30

Motivation (cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Figure: What’s in front of car? 4 4 https://3c1703fe8d.site.internapcdn.net/newman/csz/news/800/2013/aminicameras.jpg Sachin Sharma – Interconnected Intelligent cars 5 / 30

Motivation (cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Figure: Rainy weather: Sensor data might not be reliable enough 5 5 https://www.drive-safely.net/wp-content/uploads/2017/04/the-best-vehicle- dash-camera-for-poor-weather-conditions.jpg Sachin Sharma – Interconnected Intelligent cars 6 / 30

Motivation (cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Figure: Foggy weather : Sensor data might not be reliable enough 6 6 https://www.carlightblog.com/wp- content/uploads/sites/2/2016/12/Osram_Nebel_Teaser.png Sachin Sharma – Interconnected Intelligent cars 7 / 30

Motivation (cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Figure: What’s behind the curve? 7 7 https://static.pexels.com/photos/5232/road-curve-bend-ray-of- sunshine.jpg Sachin Sharma – Interconnected Intelligent cars 8 / 30

Motivation (cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Some might wonder, autonomous vehicles shouldn’t be affected by rain or fog. ◮ Rely on sensors and line of sight range. ◮ Environmental impact: Noisy sensor measurements [12]. ◮ Ghost vehicles [12] Sachin Sharma – Interconnected Intelligent cars 9 / 30

Approach Introduction Motivation Approach Features Applications Challenges Conclusion Interconnected intelligent cars can be achieved by use of : ◮ Vehicle to Vehicle Communication(V2V) ◮ Vehicle to Infrastructure Communication(V2I) Figure: V2I and V2V working in collaboration 8 8https://www.researchgate.net/profile/Brian_Smith29/publication/267206181/figure/fig2/AS:295724207099907@1447517553520/Figure- 251IllustrationofV2VandV2I16.png Sachin Sharma – Interconnected Intelligent cars 10 / 30

Approach(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion V2V solutions [13]: ◮ The ad-hoc based IEEE 802.11p standard(DSRC) ◮ Based on carrier sense multiple access with collision avoidance (CSMA/CA) ◮ Infrastructure-based cellular technologies: long term evolution (LTE). ◮ Device-to-device communication(D2D) [14] Sachin Sharma – Interconnected Intelligent cars 11 / 30

Approach(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Two techniques can be employed based on range of communication ◮ Single hop Figure: Singlehop Communication [2] A can communicate with next node/car only Sachin Sharma – Interconnected Intelligent cars 12 / 30

Approach(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion ◮ Multi hop Figure: Multihop Communication [2] B can transmit information to C, C can transfer information to A. Hence, A receives information from B. Sachin Sharma – Interconnected Intelligent cars 13 / 30

Features Introduction Motivation Approach Features Applications Challenges Conclusion The main features or data that intelligent cars share(based on DSRC [10]): ◮ Speed of car ◮ Signal timing ◮ Slippery road warning ◮ Direction of heading ◮ Messege Id Sachin Sharma – Interconnected Intelligent cars 14 / 30

Features (cont.) Introduction Motivation Approach Features Applications Challenges Conclusion ◮ Break status Figure: Break status: hard braking ahead 9 9 https://group.renault.com/wp-content/uploads/2017/06/renault- innovation-scoop-car-braking-communication.jpg Sachin Sharma – Interconnected Intelligent cars 15 / 30

Features(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion ◮ Blind spot warning Figure: Blind spot warning 10 10 https://sumeetchhawri.files.wordpress.com/2015/07/v2x_app.png Sachin Sharma – Interconnected Intelligent cars 16 / 30

Features(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Collision avoidance and security maintenance Click! 11 11 https://www.youtube.com/watch?v=WpjaEcG6I-4 Sachin Sharma – Interconnected Intelligent cars 17 / 30

Features(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Other features that can be shared: ◮ Car location [16] ◮ Path history [11] ◮ Path Indication [17] ◮ Emergency conditions: road clearance 12 12 http://sites.ieee.org/connected-vehicles/2015/09/30/first-toyota-cars-to-include- v2v-and-v2i-communication-by-the-end-of-2015/ Sachin Sharma – Interconnected Intelligent cars 18 / 30

Features(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Case Study: 2 cars going in opposite direction on a single lane Problem case scenario [17] Click! 13 13 https://www.youtube.com/watch?v=KTBdUDBceUk Sachin Sharma – Interconnected Intelligent cars 19 / 30

Features(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Solution using coordination of vehicles for the case [17] Click! 14 14 https://www.youtube.com/watch?v=KTBdUDBceUk Sachin Sharma – Interconnected Intelligent cars 20 / 30

Applications Introduction Motivation Approach Features Applications Challenges Conclusion ◮ Accident chances are reduced. ◮ Traffic jams can be avoided ◮ Parking assistance ◮ Platooning [15] ◮ Any problem occurred in one can be informed to other Sachin Sharma – Interconnected Intelligent cars 21 / 30

Challenges Introduction Motivation Approach Features Applications Challenges Conclusion Figure: Falsify identity, position or speed 15 15http://slideplayer.com/slide/6643602/23/images/45/Attack+3:+Cheating+with+Identity,+Position+or+Speed.jpg Sachin Sharma – Interconnected Intelligent cars 22 / 30

Challenges(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion Few Challenges for Interconnected intelligent cars: ◮ Security & Privacy concerns ◮ Any small fault can lead to total disturbance of driving. ◮ Some times signals might not be correct. ◮ Tests are done in controlled environment ◮ Requires infrastructure help sometimes. Sachin Sharma – Interconnected Intelligent cars 23 / 30

Challenges(Cont.) Introduction Motivation Approach Features Applications Challenges Conclusion ◮ LTE based V2V is still under standardization [9] ◮ Mixing with non-autonomous vehicles: ◮ 2017 Cadillac CTS: V2V communication ◮ 2017 Mercedes E-Class: V2X communication ◮ 2018 Mercedes S-Class: V2X communication ◮ Channel congestion and interference [10] ◮ Likely cost for a V2V system to the consumer at approximately $341- $350 per new vehicle in 2020 [10] Sachin Sharma – Interconnected Intelligent cars 24 / 30

Conclusion Introduction Motivation Approach Features Applications Challenges Conclusion ◮ Designing such a network is a challenging task. ◮ Security and privacy can be a major concern. ◮ Autonomous vehicles can be made more effective. ◮ V2V has already been implemented by major companies. ◮ In autonomous vehicles, As of now it’s a prototype. Sachin Sharma – Interconnected Intelligent cars 25 / 30

Thank you for listening Any questions?? Sachin Sharma – Interconnected Intelligent cars 26 / 30