Diferenças

Aqui você vê as diferenças entre duas revisões dessa página.

--- group_modelling [2009/03/19 13:42] – removed scosta
+++ group_modelling [2009/03/19 14:38] (atual) – tiago
@@ Linha 1: / Linha 1: @@
+====== Group 3: HUMAN ASPECTS ON CHANGE MODELLING ======
+===== 1. Cluster project proposal (input in cluster proposal template )  =====
+.1. Abstract 1.2. State-Of-The-Art, related literature 1.3. Work packages descriptions 1.4. Deliverables, time plan
+===== 2. Two publications (extended abstract at Ilhabela) =====
+  * **“Modeling Human and Institutional Aspects on Multi-Scale LUCC models”**
+===== 3. Identifying suitable tools for development and evaluation =====
+  * TerraLib (database, elementary topological operations)
+  * TerraHS/Haskell (prototype implementation)
+  * Secondo (database for moving objects),
+  * Aguila (visualisation)
+  * TerraME (environmental modelling)
+  * PCRaster (environmental modelling)
+  * TerraView (wants to do animation)
+  * R, aRT
+  * OGC web services – 52N suite, ...
+===== 4. Two topics for jointly supervised Masters- and PhD theses =====
+==== "An agent architecture for simulating human behavior in environmental change models" ====
+ Besides humans and institutions are the major driver of environmental changes [Lambin, Turner et al. 2001; Parker, Berger et al. 2001], models which take in consideration their motivations and the way they take decisions or reason about the environment they are embedded are yet in a very early stage. Several agent theories [Wooldbridge, Jennings 1995] and architectures [Minar 1996; Swarm 2008; NetLogo 2008, Repast 2008] have been proposed on the literature. However, they were not designed for simulation of human or institutional activities in environmental systems. Therefore, they lack on meet the whole requirements of this application.
+The Nested-CA model of computation, implemented in the TerraME software, has been designed for multiple scale environmental change model development [Carneiro 2006; Moreira 2008; Carneiro 2008]. However, TerraME provides just a situated view of control in which agents can acts based on their correlated state with the environment state, without the needing for explicit knowledge representation or symbolic processing. This approach supposes that at the model design time, the modeler is capable of enumerate all the possible states of the environment and provide the reaction of each agent to each of these states [Rosenschein, Kaelbling 1995]. Besides [Maes 1990] and [Fasciano 1994;  Devigne, D., Mathieu, P., Routier, J.C 2004] have shown that situated agents can have goals and plan, there are no evidences that TerraME agents can do that. Moreover, TerraME has just a very simple remote method invocation mechanism for agent communication that has yet not been used for collaborative agent modeling.
+This way, the following issues should be addressed in this work in order to provide an agent architecture for simulating human and institutional behavior in environmental change models: What are the requirements for such agent architecture? How agents can autonomously take decisions? How to simulate goal oriented behavior? How to simulate knowledge based decision taking process? How agents can plan their actions? How to simulate collaborative and competitive behavior?
+**References:**
+    * Carneiro, T. G. S., Câmara, G., Maretto, R. V., (2008) "Irregular Cellular Spaces: Supporting Realistic Spatial Dynamic Modeling over Geographical Databases". X Brazilian Symposium on Geoinformatics (GeoInfo 2008). Rio de Janeiro, Brazil .
+    * Carneiro, T. G. S.. (2006) “Nested-CA: a foundation for multiscale modeling of land use and land change”. PhD Thesis in Computer Science. Sao Jose dos Campos: INPE, 2006. Computer Science Department.
+    * Devigne, D., Mathieu, P., Routier, J.C. (2004) “Planning for spatially situated agents” Proceedings. IEEE/WIC/ACM International Conference on Intelligent Agent Technology (IAT 2004). Volume , Issue , 20-24 Sept.Page(s): 385 – 388
+    * Fasciano, M. (1994) “Situated agents can have plans”.  Proceedings of the twelfth national conference on Artificial intelligence (vol. 2), American Association for Artificial Intelligence,   Menlo Park, CA, USA.
+    * Lambin, E. F., B. L. Turner, et al. (2001). "The causes of land-use and land-cover change: moving beyond the myths." Global Environmental Change-Human and Policy Dimensions 11(4): 261-269.
+    * Maes, P. (1990). “Situated Agents Can Have Goals”. Robotics and Autonomous Systems 6: 49-70.
+    * Minar, N., Burkhart, R., Langton, C., Askenazi, M. (1996) “The Swarm simulation system: A toolkit for building mult-agent simulations”. Working Paper 96-06-042, Santa Fe Institute, Santa Fe.
+    * Moreira, E.; Aguiar, A. P.; Costa, S.; Câmara, G., (2008), Spatial relations across scales in land change models, in Vinhas, L., ed., X Brazilian Symposium on Geoinformatics, Rio de Janeiro, Brazil.
+    * Parker, D. C., T. Berger, et al. (2001). “Agent-Based Models of Land-Use and Land-Cover Change.” Report and Review of an International Workshop. L. R. No.6. Irvine, California, USA.
+    * Rosenschein, S. J., L. P. Kaelbling (1995). "A situated view of representation and control." Artificial Intelligence 73(149-73).
+    * Wooldbridge, M.; Jennings, N. R, (1985). Agent theories, architectures, and languages: a survey. Proceedings of the workshop on agent theories, architectures, and languages on Intelligent agents, Amsterdam, The Netherlands.
+    * Swarm. Swarm Development Group. http://repast.sourceforge.net/repast_3/index.html. Last access: July/2008.
+    * NetLogo. http://ccl.northwestern.edu/netlogo/. Last access: July/2008.
+    * Repast. Repast Agent Simulation Toolkit. http://repast.sourceforge.net/repast_3/index.html. Last access: July/2008.
+\\
+==== "Complex Network Modeling Support" ====
+Land changes are results from complex social and biophysical systems and their interactions (Turner et al., 1995). Such interactions result from processes that act on different hierarchical levels. At the global scale, the national and international commodities market (beef, grains and timber) drives demand for land change. This way, these changes cannot be adequately understood without knowing their linkages to decisions and structures made elsewhere. In this sense, understanding the role of networks is essential to understanding how to decide about the land use.
+A network is a set of items, which we will call vertices or nodes, with connections between them, called edges. Networks exist everywhere and at every scale and can be physical, such as infrastructure networks, and logical ones, such as market chains, linking a certain location to distant consumption or influential sites. Several works argue that networks are important tool to describe complex systems, REFS. <gpm>
+We believe which networks can be a important tool to link the actor of land change to global scenarios, as market and police public in land change modeling. However, is necessary to solve some specific challenges:
+  * Network Architecture, including data structures and algorithms.
+=== References ===
+  * Aguiar, A., Câmara, G., Cartaxo., R.: Modeling Spatial Relations by Generalized Proximity Matrices. V Brazilian Symposium  in Geoinformatics  - GeoInfo 2003, Campos do Jordão, SP, Brazil (2003)
+  * Andrade, P. R.  (2008). Modelling and  simulation  of  complex  systems with  geospatial agents.
+  * Barabási,  A.  L.,  &  Albert,  R.  (1999).  Emergence  of  scaling  in  random  networks.Science , 286, pp. 509-512.
+  * Barabási, A. L., & Bonabeau, E. (2003). Scale-free networks. Scientific American , 288,50-59.
+  * Newman, M., Barabási, A. L., & Watts, D.  J.  (2006). The  structure and dynamics of networks. EUA: Princeton University press.
+  * Moreira, E., Aguiar, A.P., Costa, S., Câmara, G.: Spatial relations across scales in land  change  models.  In:  Vinhas,  L.  (ed.):  X  Brazilian  Symposium  on Geoinformatics, GeoInfo 2008. SBC, Rio de Janeiro (2008)
+  * Watts, D. J. (1999). Small worlds. EUA: Princeton University Press.
+===== 5. One mobility measure proposal =====
+Why somebody should go? Why somebody should come?
+===== 6. Presentation of preliminary results (15+15 min. each) on Tue, Mar 17, 13.30-15.30 =====
+===== 7. Presentation of final results (15+15 min. each) on Thu, Mar 19, 14.00-16.00 =====