M1: Specifications document (01-08-2019)
Specifications of the most relevant use cases, task outputs, functionalities of the target prototype and key performance indicators to guarantee that contributions are developed in accordance with a well-defined set of requirements and design principles.
M2: Integrative data repository (01-01-2020)
Multi-dimensional data repository with updating and cleaning facilities that consolidates all available data sources with potential impact on city traffic analysis (including multi-source urban data and external data sources for gathering context information).
M3: Context-sensitive multi-source learning algorithms (01-05-2021)
Advanced algorithms able to learn descriptive and predictive models from heterogeneous urban data in the presence of situational context data; including spatiotemporal pattern miners to detect non-trivial traffic dynamics and deep learning methods to anticipate mobility problems.
M4: Traffic flow control system (01-09-2021)
Networks of reinforcement learning agents able to optimize traffic flow and adapt to context information; therefore combining state-of-the-art architectures for traffic simulation with richer reinforcement learning models based on recent advances on deep reinforcement learning.
M5: Decision support system for mobility optimization (01-12-2021)
Fully-functional prototype integrating descriptive, predictive and prescriptive algorithms made available in the two previous milestones; providing guarantees of adaptability, actionability and statistical significance in accordance with the specifications (Milestone 1).
Specifications of the most relevant use cases, task outputs, functionalities of the target prototype and key performance indicators to guarantee that contributions are developed in accordance with a well-defined set of requirements and design principles.
M2: Integrative data repository (01-01-2020)
Multi-dimensional data repository with updating and cleaning facilities that consolidates all available data sources with potential impact on city traffic analysis (including multi-source urban data and external data sources for gathering context information).
M3: Context-sensitive multi-source learning algorithms (01-05-2021)
Advanced algorithms able to learn descriptive and predictive models from heterogeneous urban data in the presence of situational context data; including spatiotemporal pattern miners to detect non-trivial traffic dynamics and deep learning methods to anticipate mobility problems.
M4: Traffic flow control system (01-09-2021)
Networks of reinforcement learning agents able to optimize traffic flow and adapt to context information; therefore combining state-of-the-art architectures for traffic simulation with richer reinforcement learning models based on recent advances on deep reinforcement learning.
M5: Decision support system for mobility optimization (01-12-2021)
Fully-functional prototype integrating descriptive, predictive and prescriptive algorithms made available in the two previous milestones; providing guarantees of adaptability, actionability and statistical significance in accordance with the specifications (Milestone 1).