In participatory crowdsensing, data sparsity in underrepresented areas and limited user engagement remain a persistent challenge. We present SpatioSense, a modular participatory crowdsensing platform that fully automates heterogeneous campaign prototyping, while addressing the challenge of data sparsity via hybrid task allocation models. We further leverage SpatioSense's modular architecture to enable rapid adaptation to diverse campaigns, as well as the integration of a novel context-aware task allocation mechanism, in which spatio-temporal and natural language processing models are used to match sensing tasks to participants. In this paper, we present the SpatioSense platform with its novel design features and adaptive context-aware task allocation algorithm, and our preliminary results.

Traditional graph-based ranking models that treat locations as isolated entities often fail to capture the complex spatio-temporal dependencies inherent in pickup and delivery platforms. Regardless of the quality of learning models adopted to predict future demand, their results can be further enhanced by encoding the spatio-temporal graph-based structure of the road network. We define a spatio-temporal graph-based measure for POI ranking, namely ZoneRank, that encodes spatial relationships, mobility flows, and temporal transitions in such platforms. Furthermore, we implement a multi-purpose ridesharing simulator to evaluate the effectiveness of ZoneRank in the context of idle vehicle repositioning.

A prevalent problem in current e-hailing and ride sharing platforms is deadheading; i.e., drivers roaming around already busy streets in search for rides, which increases congestion in urban areas; leading to negative impact on the environment, as well as the driver and customer experience. In this paper, we define a repositioning and routing algorithm for idle vehicles in urban settings that can be coupled with existing demand forecasting models. We analyze and evaluate our algorithms using benchmarks with real ride traces from NYC and San Francisco, that highlight their strength in reducing deadheading while improving drivers’ average income and not sacrificing customer pickup delays. Moreover, in areas with sparse demand, they achieve reductions in ride pickup delays and deadheading up to 40%.

While many countries are developing appropriate actions towards a greener future and moving towards adopting sustainable mobility activities, the real-time management and planning of innovative transportation facilities and services in urban environments still require the development of advanced mobile data management infrastructures. Novel green mobility solutions, such as electric, hybrid, solar and hydrogen vehicles, as well as public and gig-based transportation resources are very likely to reduce the carbon footprint. However, their successful implementation still needs efficient spatio-temporal data management resources and applications to provide a clear picture and demonstrate their effectiveness. This paper discusses the major data management challenges, open issues, and application opportunities closely related to urban green mobility. Additionally, it reports on recent successful experiences and challenging research questions. Furthermore, it highlights the global benefits one can expect when developing green mobility and emphasizes how mobile data infrastructures and services will play a crucial role in achieving these goals.

Recently ride-sharing platforms have struggled with a decreased supply of drivers, which has negatively impacted their passengers, by subjecting them to long delays and extremely high surge prices. An approach for mitigating these problems is for service providers to facilitate and coordinate carpooling via the recommendation of individually curated paths, not necessarily the shortest, for drivers towards completing their chosen rides. In this paper, we redesign the Weight Evolving Temporal graph structure to efficiently encode large dynamic road networks with temporal ride availability. Leveraging that graph structure, we efficiently define a polynomial-time optimal route recommendation algorithm that increases carpooling opportunities, taking into consideration the spatio-temporal constraints of both drivers and rides in such a highly-dynamic setting. Finally, we use simulations to demonstrate the effectiveness of these route recommendations, on both the driver and passenger experience.

With the recent trend in Mobile Crowd Sensing (MCS), i.e., using the power of crowds to assist in completing spatio-temporal sensory tasks, the pool of resources suitable for sensor systems has expanded to include already roaming devices. In this work, we present a model of MCS, in which agents share their journey information and allow the platform to guide them through their journey, completing spatio-temporal tasks on their way, in return for monetary rewards. In this paper, we formulate the task allocation problem as a routes coordination problem for participating agents. We define an optimal routing algorithm for a single agent, with an objective to maximize the rewards collected from performing tasks, which is used to define a 1/2-approximation algorithm to coordinate the routes of multiple agents. The algorithm is accompanied with an incentive compatible, rational, and cash-positive payment mechanism, which guarantees that an agent's truthful participation is an ex-post Nash equilibrium strategy, in an optimal setting. Finally, we analyze the defined mechanisms theoretically, and evaluate their performance experimentally using real mobility traces from urban environments.