Dr Sebastian Stein is an Associate Professor in the Agents, Interaction and Complexity (AIC) research group at the University of Southampton. He is an expert in designing algorithms and incentive mechanisms for systems where humans and intelligent software agents collaborate. He has applied his expertise to future transportation systems, the smart grid, crowdsourcing and cloud computing. Dr Stein has over 50 peer-reviewed publications in top artificial intelligence conferences and journals, including AIJ, JAIR, IJCAI, AAAI and AAMAS. He leads the Services theme within the AutoTrust project.
Services Theme: As vehicles share data over the IoV, there is great potential for new service offerings to emerge. Vehicles can organise into platoons on motorways to save fuel, they can coordinate their travel plans to avoid congestion at bottlenecks or drivers can offer ridesharing opportunities to others. When choosing what services to participate in, human drivers or passengers will need to consider how the terms of that service align with their values, and what the benefits and costs of participation are. Designing effective services will therefore require both deep understanding of how to design (1) privacy-aware and (2) incentive-aware services. Regarding the first, services must flexibly deal with consumers that are willing to share more or less data. In our human-centred approach, our focus is to privilege privacy by design: services where possible need to work on anonymous, aggregated data; where personal data is required, key decisions (e.g., choosing routes or departure times) should be made on the consumer’s personal and trusted devices rather than on a centralised and monolithic server. The second and complementary challenge acknowledges that often travelers need incentives to change their behaviour patterns e.g., to join and lead a platoon, to avoid peak times or to switch to public transport to reduce congestion. There is considerable work in how incentives can work to increase the efficiency of the overall system, but to deliver these approaches within the complex and dynamic system that is the IoV, taking into account the personal preferences of travelers, is a new and complex domain space.