http://www.dia.fi.upm.es/~jdlope/ Javier de Lope RSS Feed http://www.dia.fi.upm.es/~jdlope/publications.html#journals It is available the online publication of a new paper on the self-emergence of a common lexicon in teams of autonomous agents in Neurocomputing (Elsevier). Tue, 20 Sep 2011 13:31:49 GMT http://www.dia.fi.upm.es/~jdlope/conferences.html We have proposed a Special Session on Emergent Applications of Natural Computing Systems in the IEEE NaBIC 2011. Mon, 20 Jun 2011 16:00:25 GMT http://www.dia.fi.upm.es/~jdlope/publications.html#book-chapters We safely returned from the IWINAC 2011 where we met with our friends and colleagues and also presented our latest works. Mainly they deal on multi-robot systems. Feel free to review them. There also a couple of recent references in the International Conferences section. Mon, 6 Jun 2011 13:21:02 GMT http://www.dia.fi.upm.es/~jdlope/publications.html#journals We propose an adaptive neuron-like network based on learning automata and granular inference systems. We apply it to Pattern Recognition and Machine Learning. The publication is available in Neurocomputing (Elsevier). Mon, 14 Mar 2011 17:19:55 GMT http://www.dia.fi.upm.es/~jdlope/publications.html#journals The aim of this paper is to present a series of robust high performance implementations of reinforcement learning algorithms based on temporal-difference learning and weighted k-nearest neighbors for linear function approximation. The publication is available in Neurocomputing (Elsevier). Mon, 14 Mar 2011 17:17:03 GMT http://www.dia.fi.upm.es/~jdlope/publications.html#journals It is available the Special Issue on Intelligent Robotics and Neuroscience in the Robotics and Autonomous Systems journal (Elsevier). Fri, 28 Jan 2011 09:20:17 GMT http://www.dia.fi.upm.es/~jdlope/conferences.html Proposed a Special Sesion on Hybrid Intelligent Approaches in Cooperative Multi-robot Systems. See the abstract for a description and topics. Fri, 17 Nov 2010 09:36:58 GMT http://www.youtube.com/watch?v=Ee_jBnS2s_0 This video corresponds to the final work of one of my students (Raul Dominguez) in an experimental course on Artificial Intelligence and Robotics. The assignments correspond to the construction of three modules: a navigation module which provides acting and low-level perception, a planning module concerned with knowledge representation and search, and lastly a vision module to detect and recognize markers. Fri, 19 Nov 2010 15:01:49 GMT http://www.dia.fi.upm.es/~jdlope/conferences.html Proposed a Special Sesion on Multi-robot Systems. See the abstract for a description and a non-exaustive list of tentative topics. Fri, 19 Nov 2010 11:57:35 GMT http://www.dia.fi.upm.es/~jdlope/publications.html#book-chapters Updated the publications with a reference to our latest work on evolutionary communication in a population of autonomous agents. Tue, 28 Sep 2010 10:14:49 GMT http://www.dia.fi.upm.es/~jdlope/programming.html It is time to consider if Python can be a good alternative to Matlab. Advantages? Mon, 06 Sep 2010 12:07:42 GMT http://www.dia.fi.upm.es/~jdlope/programming.html Lisp is faster than many other languages but the running time is not the only issue. Fri, 12 Apr 2010 21:52:27 GMT http://www.dia.fi.upm.es/~jdlope/links.html The index gives an indication of the popularity of programming languages. There are very interesting info there. Fri, 9 Apr 2010 09:42:09 GMT http://www.dia.fi.upm.es/~jdlope/programming.html Some experiments on reinforcement learning and the gridworld. There are some curious results by using and comparing three different programming languages. Sun, 4 Apr 2010 19:05:21 GMT http://www.dia.fi.upm.es/~jdlope/ New RSS feed. The channel informs about the latest website updates and also about YouTube videos and general news. Wed, 31 Mar 2010 13:31:28 GMT http://www.youtube.com/watch?v=NM4QoXNLUTQ Two robots build a metric map of its environment. We use Occupancy Grids (the Histogram method) for saving the sonar readings. Each robot creates a zone of the world model in a shared map. We try to check if the performance of this task can be improved by using a robot colony. Sun, 28 Mar 2010 20:04:27 GMT http://www.dia.fi.upm.es/~jdlope/ I've updated the style sheet colors and background. Now it is even simpler. I've also fixed some errors and warnings. Now the website is validated as XHTML 1.0 Strict and the style sheet as CSS 2.1 Sun, 28 Mar 2010 18:51:29 GMT http://www.youtube.com/watch?v=7bRlznnPH6E We use a couple of Lego NXT robots for implementing this behavior. The leader robot has an infrared beacon. It is controlled by hand. We use a Wiimote to determine the command that must be sent to the robot. The follower robot detects the infrared beacon. We are also using a Wiimote. Now the Wiimote is aboard the robot. The Wiimote can detect and track until 4 infrared sources. All these signals are used for implementing the behavior. Sun, 28 Mar 2010 17:47:29 GMT http://www.youtube.com/watch?v=7OLCAk6fdOM The goal of the Project H1 is to control a conventional radio controlled helicopter by means of an external camera. The images are processed in order to obtain the helicopter coordinates in the image framework. Those coordinates are associated to the sticks movements in the radio controller. We use an time-delay recurrent artificial neural network for modeling the flight dynamics. The video shows some tracking tests. Some flight parameters such as helicopter coordinates and its estimated speed are shown in top left. The real radio controller stick positions are shown in bottom right. Mon, 22 Mar 2010 17:43:14 GMT http://www.youtube.com/watch?v=DLuPSRF-Ucc Simple collision avoidance behavior for checking out the e-puck robot in the Webots simulator. Both front distance sensors are used in order to detect the obstacles. When an obstacle is detected, the robot turns left or right. The direction is determined by using the lateral sensors. Mon, 15 Mar 2010 09:36:54 GMT