{"title":"Legs that can walk: embodiment-based modular reinforcement learning applied","authors":"D. Jacob, D. Polani, Chrystopher L. Nehaniv","doi":"10.1109/CIRA.2005.1554304","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554304","url":null,"abstract":"Experiments to illustrate a novel methodology for reinforcement learning in embodied physical agents are described. A simulated legged robot is decomposed into structure-based modules following the authors' EMBER principles of local sensing, action and learning. The legs are individually trained to 'walk' in isolation, and re-attached to the robot; walking is then sufficiently stable that learning in situ can continue. The experiments demonstrate the benefits of the modular decomposition: state-space factorisation leads to faster learning, in this case to the extent that an otherwise intractable problem becomes learnable.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129220571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Integrating computer vision libraries in networked robotic systems","authors":"E. Cervera","doi":"10.1109/CIRA.2005.1554288","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554288","url":null,"abstract":"The niche where computer vision software meets networked robotics remains unfilled so far. Descending costs, and technological advances in wireless communications, digital video capture, and computing power provides roboticists with powerful yet affordable tools to develop cooperative, distributed, vision-based control tasks. This paper presents a consistent framework for such networked, visually-controlled environments, where the main assumption is that video may be captured, processed, and used for control in different machines across the network. To do so, care has been taken in order to ensure efficiency, scalability, and reusability of software. The presented framework aims to seamlessly integrate rock-solid, fast, existing vision processing libraries within a cross-platform, distributed environment, based on Java and agents. Different applications in real setups are envisioned.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115458851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Computational intelligence for high-precision industrial robots","authors":"S. Moberg","doi":"10.1109/CIRA.2005.1554244","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554244","url":null,"abstract":"Summary form only given. ABB has launched its fifth generation robot controller, the IRC5. The basic motion control of the IRC5 controller includes the TrueMove and QuickMove functionality for best possible production rate and quality. In addition to this, the MultiMove function enables new production concepts and more cost-effective solutions. To penetrate new market segments, ease-of-use, high precision and cost-effectiveness are important factors. The availability of low-cost computational power and, in some cases, sensor integration enables the use of cost-effective industrial robots for high-precision tasks.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126148089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A wearable inertial sensor system for human motion analysis","authors":"Tao Liu, Y. Inoue, K. Shibata, Xueyan Tang","doi":"10.1109/CIRA.2005.1554311","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554311","url":null,"abstract":"To analyze the multi-dimension motion of human foot during walking, a wearable inertial sensor system is proposed. In this system, two gyroscopes (ENC-05EB) were used to measure angular velocities in x- and y-directions, and an accelerometer (ADXL202) was used to measure acceleration in x- and y directions respectively. Furthermore, a fuzzy inference system (FIS) was developed for the calculation of the gait phases derived from sensors' outputs. A digital filter was also designed to remove noises from of the output of the fuzzy inference system, which enhances robustness of the system. Finally, experimental study was conducted, and a wearable mechanism with the sensors system was constructed to detect gait phases of the normal human walking.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"300 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122009424","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Realtime estimation of human hand posture for robot hand control","authors":"K. Hoshino, T. Tanimoto","doi":"10.1109/CIRA.2005.1554261","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554261","url":null,"abstract":"The authors propose a system for searching the shape of human hands and fingers in real time and with high accuracy, without using any special peripheral equipment such as range sensor, PC cluster, etc., by a method of retrieving similar image quickly with high accuracy from a large volume of image database containing the complicated shapes and self-occlusions. In designing the system, we constructed a database in a way to be adaptable even to differences among individuals, and searched CG images of hand similar to unknown hand image, through extraction of characteristics using high-order local autocorrelational patterns, reduction of the amount of characteristics centering on principal component analysis, and prior rearrangement of data corresponding to the amount of characteristics. As a result of experiments, our system performed high-accuracy estimation of human hand shape where mean error was 7 degrees in finger joint angles, with the processing speed of 30 fps or over.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122671024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Automatic building detection from aerial images for mobile robot mapping","authors":"M. Persson, Mats Sandvall, T. Duckett","doi":"10.1109/CIRA.2005.1554289","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554289","url":null,"abstract":"To improve mobile robot outdoor mapping, information about the shape and location of buildings is of interest. This paper describes a system for automatic detection of buildings in aerial images taken from a nadir view. The system builds two types of independent hypotheses based on the image contents. A segmentation process implemented as an ensemble of SOMs (Self Organizing Maps) is trained and used to create a segmented image snowing different types of roofs, vegetation and sea. A second type of hypotheses is based on an edge image produced from the aerial photo. A line extraction process uses the edge image as input and extracts lines from it. From these edges, corners and rectangles that represent buildings are constructed. A classification process uses the information from both hypotheses to determine whether the rectangles are buildings, unsure buildings or unknown objects.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123413337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An adaptive task assignment method for multiple mobile robots via swarm intelligence approach","authors":"Dandan Zhang, G. Xie, Junzhi Yu, Long Wang","doi":"10.1109/CIRA.2005.1554312","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554312","url":null,"abstract":"This paper describes an adaptive task assignment method for a team of fully distributed mobile robots with initially identical functionalities in unknown environments. The method is applicable for mediate- to large-scaled robot groups and tasks. A hierarchical architecture for task assignment is established for each individual robot. In the higher hierarchy, the self-reinforcement learning model inspired by the behaviors of social insects is employed to differentiate the initially identical robots into different kinds of high-level task \"specialists\"; while in the lower hierarchy, ant system algorithm is adopted to organize low-level task assignment. To avoid using a centralized component, \"local blackboard\" communication mechanism is utilized for knowledge sharing. The proposed method allows the robot team members to adapt themselves to the unknown dynamic environments, respond flexibly to the environment perturbations and robustly to the modifications in the team arising from mechanical failure. Simulations of a cooperative collection task validate the effectiveness of the presented method.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131469509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Robotic airship mission path tracking control based on human operator's skill","authors":"Jun Luo, Shaorong Xie, Jinjun Rao, Zhenbang Gong","doi":"10.1109/CIRA.2005.1554332","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554332","url":null,"abstract":"A yawing controller based on artificial neural networks (ANN) and human operator's skill is presented for robotic airship mission path tracking. Firstly, consideration of the path tracking errors from the point of view of operators is presented. Then, a data acquisition system is designed to collect flight data under manual control. Thirdly, The processed flight data are used to train and validate a multilayer feedforward ANN offline. Lastly, the trained ANN is reconstructed in the flight control system for yawing control. The experimental results indicate that this solution is valid and the ANN controller is robust even with wind disturbance.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123148272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Experimental comparison of a directed and a non-directed potential field approach to formation navigation","authors":"Frank E. Schneider, D. Wildermuth","doi":"10.1109/CIRA.2005.1554249","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554249","url":null,"abstract":"The field of multirobot systems (MRS) is continuously drawing more attention. The DARPA SDR program and the RoboSoccer/RobotRescue contests have shown the great potential of this field. One of the increasingly popular uses of the field of MRS is in communication and sensor-networks. These applications require a task depended navigation. The MRS sub-problem of formation navigation is predestined for this purpose. This paper presents an improvement of the classical potential field approach for motion co-ordination in formations of MRS. The approach is briefly compared to related work. For the subtask of moving in formation, a solution based on a directed potential field is presented. Different forces belonging to other robots, obstacles and the aspired shape of formation are combined and used to move each robot to its desired position inside the formation. While moving in formation, the group is able to avoid obstacles and approach towards a specified target. First experiments that show the differences and improvements in comparison to the classical method are presented.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"68 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120894807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. López-de-Ipiña, Juan Ignacio Vázquez, Jonathan Ruiz de Garibay, D. Sainz
{"title":"An architecture for sentient GPRS-enabled MicroBots","authors":"D. López-de-Ipiña, Juan Ignacio Vázquez, Jonathan Ruiz de Garibay, D. Sainz","doi":"10.1109/CIRA.2005.1554268","DOIUrl":"https://doi.org/10.1109/CIRA.2005.1554268","url":null,"abstract":"Ambient intelligence is a user-centered concept which combines several computing disciplines with the purpose of enhancing/facilitating the user's daily activities. We deem that autonomous or semiautonomous (remotely controlled) sentient MicroBots may also be first-class citizens within ambient intelligence. Those MicroBots would interact with their surrounding environment assisted by their built-in sensors, effectors and communication facilities, on behalf of the users they serve. In essence, they would also profit as users do from Ambient Intelligence to achieve their programmed goals. In this paper, we contribute with a solution to enable the realtime remote control of GPRS-accessible semiautonomous sentient MicroBots. Moreover, we discuss the extensions necessary to convert those MicroBots into care assistants for disabled and elderly people.","PeriodicalId":162553,"journal":{"name":"2005 International Symposium on Computational Intelligence in Robotics and Automation","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114974146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}