World wide transport authorities are imposing complex Hours of Service (from now on, HoS) regulations to drivers (Meyer 2011, Goel and Vidal 2013), which constraint the amount of working, driving and resting time when delivering a service. As a consequence, transport companies are responsible not only of scheduling driving plans aligned with laws that define the legal behaviour of a driver, but also of monitoring and identifying as soon as possible problematic patterns that can incur in costs due to sanctions. Fortunately, the widespread adoption of onboard IoT devices in vehicle fleets enables recording of the driver activities in event logs, but the large amount of data ingested makes difficult for transport experts to understand what happened and to make actions that forestall illegal behaviour. For this reason, an important technical challenge is to come up with easily interpretable descriptive models that help understand the huge amount of information stored in such event logs. The main objective not only consists of finding out if drivers workplan complies with the HoS regulation, but also summarising their activities in a concise but representative way. Additionally, these underlying patterns in the event log could be analysed in order to discover driving styles which could make possible the suggestion of routes or tasks more aligned to the driver preferences. The creation of driver profiles based on driving styles with HoS can be extremely useful for managers, as they could assign transport routes to the most appropriate drivers, given the length of the route and the proximity of the deadline. For example, drivers who maximise their driving hours could be preferred for long distance routes and drivers who tend to take split rest to on-city deliveries.

This project proposes a methodology for the automatic generation of action models from video game dynamics descriptions, as well as its integration with a planning agent for the execution and monitoring of the plans. Planners use these action models to get the deliberative behaviour for an agent in many different video games and, combined with a reactive module, solve deterministic and no-deterministic levels. Experimental results validate the methodology and prove that the effort put by a knowledge engineer can be greatly reduced in the definition of such complex domains. Furthermore, benchmarks of the domains has been produced that can be of interest to the international planning community to evaluate planners in international planning competitions.

In this paper we describe the integration of a robot control platform (Player/Stage) and a real robot (Pioneer P3DX) with PELEA (Planning, Execution and LEarning Architecture). PELEA is a general-purpose planning architecture suitable for a wide range of real world applications, from robotics to emergency management. It allows planning engineers to generate planning applications since it integrates planning, execution, replanning, monitoring and learning capabilities. We also present a relational learning approach for automatically modeling robot-action execution durations, with the purpose of improving the planning process of PELEA by refining domain definitions.