This project study and develop behavior based mobile robots. Initially, the project focused on developing new algorithm and methods for mobile robot applications with reduced sensory capabilities or with low cost sensors and actuators. The principal methods that we use include genetic algorithms, genetic programming and fuzzy logic.

Since 2007, the project study state of art algorithms and architectures to determine capabilities and limitations in real world setups. In order to make this process more efficient we divide the process into two stages. In the first stage, we utilize mobile robot simulators. In the second stage, we adapt and develop new algorithms capable of completing tasks in real world scenarios.

The project was started by Prof. Tomás Arredondo, Nicolás Navarro, César Muñoz on the second half of 2005 after a seminar on artificial intelligence. For more information, please visit the project official web site here.

Example of different trajectories generated by our genetic algorithms and fuzzy motivations approach. Left, trajectories displayed in YAKS. Right, trayectories displayed using scilab.

Left, IRMA-II, our mobile robot platform. Right, Real-world results recorded by a ceilling camera

Related Publications:

1. Fuzzy Motivations in a Multiple Agent Behaviour-Based Architecture
   International Journal of Advanced Robotic Systems. vol. 10, no. 313, pp. 1-13, Aug 2013
   Arredondo, Tomás; Freund, Wolfgang; Navarro-Guerrero, Nicolás; Castillo, Patricio
   doi, url, Copyright (©) 2013 Arredondo et al., PDF, bibtex, key: Arredondo2013Fuzzy,
2. Desarrollo e implementación de algoritmo evolutivo multi-objectivo para generar rutas online en un robot móvil autónomo
   Universidad Técnica Federico Santa María. Valparaíso, Chile, Dec 2009 Language: Spanish
   Navarro-Guerrero, Nicolás
   url, Copyright (©) 2009 Navarro-Guerrero, bibtex, key: Navarro-Guerrero2010Desarrollo,
3. Cooperative Adaptive Behavior Acquisition in Mobile Robot Swarms Using Neural Networks and Genetic Algorithms
   Electronics, Robotics and Automotive Mechanics Conference (CERMA). pp. 417-421, Morelos, Mexico, Sep 2008
   Muñoz, César; Navarro, Nicolás; Arredondo, Tomás; Freund, Wolfgang
   doi, url, ©2008, IEEE, bibtex, key: Munoz2008Cooperative,
4. Real-Time Adaptive Fuzzy Motivations for Evolutionary Behavior Learning by a Mobile Robot
   Mexican International Conference on Artificial Intelligence (MICAI) - Advances in Artificial Intelligence. vol. 4293 of LNCS, pp. 101-111, Apizaco, Mexico, Nov 2006
   Freund, Wolfgang; Arredondo Vidal, Tomás; Muñoz, César; Navarro, Nicolás; Quirós, Fernando
   doi, url, ©2006, Springer-Verlag Berlin Heidelberg, PDF, bibtex, key: Freund2006Real,
5. Acquiring Adaptive Behaviors of Mobile Robots Using Genetic Algorithms and Artificial Neural Networks
   Electronics, Robotics and Automotive Mechanics Conference (CERMA). vol. 1, pp. 87-91, Cuernavaca, Mexico, Sep 2006
   Navarro, Nicolás; Muñoz, César; Freund, Wolfgang; Arredondo, Tomás
   doi, url, ©2006, IEEE, PDF, bibtex, key: Navarro2006Acquiring,
6. Fuzzy Motivations for Evolutionary Behavior Learning by a Mobile Robot
   International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems (IEA/AIE). vol. 4031 of LNCS, pp. 462-471, Annecy, France, Jun 2006
   Arredondo, Tomás; Freund, Wolfgang; Muñoz, César; Navarro, Nicolás; Quirós, Fernando
   doi, url, ©2006, Springer-Verlag Berlin Heidelberg, PDF, bibtex, key: Arredondo2006Fuzzy,
7. A Neural Approach for Robot Navigation Based on Cognitive Map Learning
   International Joint Conference on Neural Networks (IJCNN). pp. 1146-1153, Brisbane, QLD, Australia,
   Yan, Wenjie; Weber, Cornelius; Wermter, Stefan
   doi, url, ©2012, IEEE, key: Yan2012Neural,

---