EKEMAS, an agent-based geo-simulation framework to support continual planning in the real-word

Abstract

In this paper, we propose an agent-based geosimulation framework EKEMAS to assist human planners when planning under strong spatial constraints in a real large-scale space. The approach consists in drawing a parallel between the real environment (for example, a forest in fire) and the simulated environment based on GIS data. This virtual environment uses software agents which are aware of the space and equipped with advanced spatial reasoning capabilities. In addition, we suggest some enhancements for the Continual Planning approach. Our aim is to demonstrate how EKEMAS, when coupled with a continual planning approach and agent's spatial reasoning capabilities, can assist human planners overcoming obstacles related to real world constraints: dynamic, uncertain, and spatially constrained environment. We illustrate this idea on the forest firefighting problem and we use MAGS as a simulation platform and Prometheus as a fire simulator. Finally, and since plans in the studied case (wildfire fighting) are mainly paths, we also propose a new approach based on agent geo-simulation in order to solve particular Pathfinding problems.

Sensor Network and GeoSimulation: Keystones for Spatial Decision Support Systems

“…Indeed, when based on realistic assumptions, simulation offers an efficient way of modeling crisis response tasks. In the domain of forest firefighting for example, we developed a geosimulation-based SDSS which relies on a four-layer architecture (Sahli and Moulin, 2009;Sahli and Jabeur, 2011). An adaptation to this architecture is discussed later (see Subsection 4.3).…”

“…Static sensors are used to monitor the environment and send sensed data to the SDSS, which may use this data in different ways. For example it can map the data (usually after treatment) on the geo-referenced map (Sahli and Moulin, 2009). A human user can visualise the data before taking any decision.…”

“…The sensed data can also be used to feed a simulated model (within the SDSS) of the event occurring in the dynamic environment. We have already followed this approach for the forest fire problem in (Sahli and Moulin, 2009) and the train derailment problem (caused by rock falls) in (Sahli et al, 2008). In this section we are going to focus on another type of sensors which is very useful during crisis response but much more difficult to manage.…”

“…2. This design philosophy is inspired by the layered simulation model we proposed in (Sahli and Moulin, 2009). The real sensors are of course located in the real world.…”

“…Even a high traffic of exchanged messages will not really affect the performance of the relocation process; (ii) Since the relocation planning will be conducted by the representative agents in the SDSS, each agent will have access to all relevant data (spatial data of the surrounding space from GIS, the last evolution of the situation from the simulated environment, and the position of the other resources and sensors from the other co-existing agents). More details about the internal architecture of these agents can be found in (Sahli and Moulin, 2009); (iii) As all agents are in the SDSS, a human decision-maker can have a global idea about the current situation. Therefore, he/she can take part in the relocation planning process by dictating strategic actions.…”