Diseño de una estrategia para la planeación de rutas de navegación autónoma de un robot móvil en entornos interiores usando un algoritmo de aprendizaje automático

dc.contributor.advisorJimenez Builes, Jovani Albertospa
dc.contributor.advisorAcosta Amaya, Gustavospa
dc.contributor.authorRamírez Bedoya, Diego Leónspa
dc.contributor.researchgroupGIDIA: Grupo de Investigación y Desarrollo en Inteligencia Artificialspa
dc.date.accessioned2021-01-18T15:51:17Zspa
dc.date.available2021-01-18T15:51:17Zspa
dc.date.issued2020-10-09spa
dc.description.abstractThe problem of autonomous robot navigation in internal environments must overcome various difficulties such as the dimensionality of the data, the computational cost and the possible presence of mobile objects. This thesis is oriented to the design of a planning strategy of routes for autonomous navigation of robots in interior environments based on automatic learning, for which it characterizes some strategies that the literature reports, The DQN machine learning algorithm is specified, to be implemented on the Turtlebot robotic platform of the Gazebo simulator. In addition, a series of Experiments changing the parameters of the algorithm to validate the strategy that shows how the robotic platform through the exploration of the environment and the subsequent exploitation of knowledge makes effective route planning. Video of Experiment can be found at https://youtu.be/5ehdh-BvY7E.spa
dc.description.abstractEl problema de la navegación autónoma de los robots en entornos internos debe superar varias dificultades como la dimensionalidad de los datos, el costo computacional y la posible presencia de objetos móviles. Esta tesis se orienta al diseño de una estrategia de planeación de rutas para la navegación autónoma de robots en entornos interiores con base en el aprendizaje automático. Para lo cual se caracteriza algunas estrategias que reporta la literatura, se especifica el algoritmo de aprendizaje automático DQN, para luego ser implementado en la plataforma robótica Turtlebot del simulador Gazebo. Además, se realizó una serie de experimentos cambiando los parámetros del algoritmo para hacer la validación de la estrategia que muestra como la plataforma robótica por medio de la exploración del ambiente y la posterior explotación de conocimiento hace una planeación de la ruta eficaz. Vídeo del experimento puede ser encontrado en https://youtu.be/5ehdh-BvY7E.spa
dc.description.additionalLínea de Investigación: Robóticaspa
dc.description.degreelevelMaestríaspa
dc.format.extent69spa
dc.format.mimetypeapplication/pdfspa
dc.identifier.citationRamírez, D. (2020) Diseño de una estrategia para la planeación de rutas de navegación autónoma de un robot móvil en entornos interiores usando un algoritmo de aprendizaje automático. Tesis de maestría en ingeniería de sistemas. Universidad Nacional de Colombia.spa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/78793
dc.language.isospaspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Medellínspa
dc.publisher.programMedellín - Minas - Maestría en Ingeniería - Ingeniería de Sistemasspa
dc.relationBedoya, D. L. R., Builes, J. A. J., & Bedoya, J. W. B. (2019). Metodología de desarrollo de software para plataformas educativas robóticas usando ROS-XP. Revista Politécnica, 15(30), 55-69.spa
dc.relation.referencesNiels Justesen y col. “Deep learning for video game playing”. En:IEEE Transactionson Games12.1 (2019), pags. 1-20spa
dc.relation.referencesJoohyun Woo y Nakwan Kim. “Vision based obstacle detection and collision risk esti-mation of an unmanned surface vehicle”. En:2016 13th International Conference onUbiquitous Robots and Ambient Intelligence (URAI). IEEE. 2016, p ́ags. 461-465spa
dc.relation.referencesHerke van Hoof. “Machine Learning through Exploration for Perception-Driven Robo-tics”. Tesis doct. Technische Universität Darmstadt, 2016spa
dc.relation.referencesMin Hyuk Woo, Soo-Hong Lee y Hye Min Cha. “A study on the optimal route de-sign considering time of mobile robot using recurrent neural network and reinfor-cement learning”. En:Journal of Mechanical Science and Technology 32.10 (2018), págs. 4933-4939spa
dc.relation.referencesUrcera I Martín y col. “Generación de trayectorias robóticas mediante aprendizaje profundo por refuerzo”. Tesis de maestría. Universitat Politècnica de Catalunya, 2018spa
dc.relation.referencesGary G Yen y Travis W Hickey. “Reinforcement learning algorithms for robotic navi-gation in dynamic environments”. En: ISA transactions 43.2 (2004), págs. 217-230spa
dc.relation.referencesTimothy P Lillicrap y col. “Continuous control with deep reinforcement learning”. En: arXiv preprint arXiv:1509.02971(2015).spa
dc.relation.referencesAnish Pandey, S Pandey y DR Parhi. “Mobile robot navigation and obstacle avoidance techniques: A review”. En: Int Rob Auto J 2.3 (2017), pág. 00022.spa
dc.relation.referencesSamuel Chenatti y col. “Deep Reinforcement Learning in Robotics Logistic Task Coor-dination”. En: 2018 Latin American Robotic Symposium, 2018 Brazilian Symposiumon Robotics (SBR) and 2018 Workshop on Robotics in Education (WRE). IEEE. 2018, págs. 326-332spa
dc.relation.referencesPablo Quintía Vidal. “Robots capaces de aprender y adaptarse al entorno a partir de sus propias experiencias”. Tesis doct. Universidade de Santiago de Compostela, 2013.spa
dc.relation.referencesPablo Escandell Montero. “Aprendizaje por refuerzo en espacios continuos: algoritmos y aplicación al tratamiento de la anemia renal”. Tesis doct. Universitat de València,2014.spa
dc.relation.referencesMarco Wiering y Martijn Van Otterlo. Reinforcement learning. Vol. 12. Springer, 2012.spa
dc.relation.referencesGregory Kahn y col. “Self-supervised deep reinforcement learning with generalizedcomputation graphs for robot navigation”. En: 2018 IEEE International Conferenceon Robotics and Automation (ICRA). IEEE. 2018, págs. 1-8spa
dc.relation.referencesAnis Koubâa y col.Robot Path Planning and Cooperation. Vol. 772. Springer, 2018.spa
dc.relation.referencesO Khatib y JF Le Maitre. “Dynamic control of manipulators operating in a complex environment”. En: On Theory and Practice of Robots and Manipulators, 3rd CISM-IFToMM Symp. Vol. 267. 1978.spa
dc.relation.referencesNils J Nilsson.The quest for artificial intelligence. Cambridge University Press, 2009.spa
dc.relation.referencesRamon Gonzalez, Marius Kloetzer y Cristian Mahulea. “Comparative study of trajec-tories resulted from cell decomposition path planning approaches”. En: 2017 21st In-ternational Conference on System Theory, Control and Computing (ICSTCC). IEEE.2017, págs. 49-54.spa
dc.relation.referencesJan Rosell y Pedro Iniguez. “Path planning using harmonic functions and probabilistic cell decomposition”. En: Proceedings of the 2005 IEEE international conference on robotics and automation. IEEE. 2005, págs. 1803-1808.spa
dc.relation.referencesNeerendra Kumar, Zoltán V ́amossy y Zsolt Miklós Szabó-Resch. “Robot path pur-suit using probabilistic roadmap”. En: 2016 IEEE 17th International Symposium on Computational Intelligence and Informatics (CINTI). IEEE. 2016, pags. 000139-000144spa
dc.relation.referencesMark Pfeiffer y col. “Reinforced imitation: Sample efficient deep reinforcement learning for mapless navigation by leveraging prior demonstrations”. En: IEEE Robotics and Automation Letters3.4 (2018), págs. 4423-4430.spa
dc.relation.referencesChen Xia y Abdelkader El Kamel. “Neural inverse reinforcement learning in autono-mous navigation”. En: Robotics and Autonomous Systems 84 (2016), págs. 1-14.spa
dc.relation.referencesBilly Okal y Kai O Arras. “Learning socially normative robot navigation behaviors with bayesian inverse reinforcement learning”. En: 2016 IEEE International Conference on Robotics and Automation (ICRA). IEEE. 2016, págs. 2889-2895.spa
dc.relation.referencesDevika S Nair y P Supriya. “Comparison of Temporal Difference Learning Algorithmand Dijkstra’s Algorithm for Robotic Path Planning”. En: 2018 Second InternationalConference on Intelligent Computing and Control Systems (ICICCS). IEEE. 2018, págs. 1619-1624.spa
dc.relation.referencesAleksandr I Panov, Konstantin S Yakovlev y Roman Suvorov. “Grid path planningwith deep reinforcement learning: Preliminary results”. En: Procedia computer science123 (2018), págs. 347-353.spa
dc.relation.referencesBoyi Liu, Lujia Wang y Ming Liu. “Life long federated reinforcement learning: a learning architecture for navigation in cloud robotic systems”. En: IEEE Robotics and Automation Letters 4.4 (2019), págs. 4555-4562.spa
dc.relation.referencesSomil Bansal y col. “Combining optimal control and learning for visual navigation in novel environments”. En: Conference on Robot Learning. 2020, págs. 420-429.spa
dc.relation.referencesDavid Luviano Cruz y Wen Yu. “Path planning of multi-agent systems in unknown environment with neural kernel smoothing and reinforcement learning”. En: Neurocomputing 233 (2017), págs. 34-42.spa
dc.relation.referencesPinxin Long y col. “Towards optimally decentralized multi-robot collision avoidance via deep reinforcement learning”. En: 2018 IEEE International Conference on Robotics and Automation (ICRA). IEEE. 2018, págs. 6252-6259.spa
dc.relation.referencesYue Jin y col. “Efficient multi-agent cooperative navigation in unknown environments with interlaced deep reinforcement learning”. En: ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE. 2019, págs. 2897-2901.spa
dc.relation.referencesLei Tai, Giuseppe Paolo y Ming Liu. “Virtual-to-real deep reinforcement learning: Continuous control of mobile robots for mapless navigation”. En: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE. 2017, págs. 31-36.spa
dc.relation.referencesAshwini Pokle y col. “Deep local trajectory replanning and control for robot navigation”. En: 2019 International Conference on Robotics and Automation (ICRA). IEEE. 2019, págs. 5815-5822.spa
dc.relation.referencesF Lachekhab y M Tadjine. “Goal seeking of mobile robot using fuzzy actor critic learning algorithm”. En:2015 7th International Conference on Modelling, Identificationand Control (ICMIC). IEEE. 2015, págs. 1-6spa
dc.relation.referencesAkram Adib y Behrooz Masoumi. “Mobile robots navigation in unknown environments by using fuzzy logic and learning automata”. En: 2017 Artificial Intelligence and Robotics (IRANOPEN). IEEE. 2017, págs. 58-63.spa
dc.relation.referencesChia-Feng Juang y Trong Bac Bui. “Reinforcement Neural Fuzzy Surrogate-Assisted Multiobjective Evolutionary Fuzzy Systems With Robot Learning Control Application”. En: IEEE Transactions on Fuzzy Systems28.3 (2019), págs. 434-446.spa
dc.relation.referencesFatemeh Fathinezhad, Vali Derhami y Mehdi Rezaeian. “Supervised fuzzy reinforce-ment learning for robot navigation”. En: Applied Soft Computing40 (2016), págs. 33-41.spa
dc.relation.referencesAlma Y Alanis y col. “Integration of an Inverse Optimal Neural Controller with Reinforced-SLAM for Path Panning and Mapping in Dynamic Environments”. En:Computación y Sistemas19.3 (2015), págs. 445-456.spa
dc.relation.referencesVignesh Prasad y col. “Learning to Prevent Monocular SLAM Failure using Reinforce-ment Learning”. En: Proceedings of the 11th Indian Conference on Computer Vision, Graphics and Image Processing. 2018, págs. 1-9.spa
dc.relation.referencesBeril Sirmaçek y col. “Reinforcement learning and slam based approach for mobile robot navigation in unknown environments”. En: ISPRS Workshop Indoor 3D 2019.2019.spa
dc.relation.referencesSiti Sendari y col. “Exploration of genetic network programming with two-stage rein-forcement learning for mobile robot”. En: Telkomnika 17.3 (2019), págs. 1447-1454.spa
dc.relation.referencesAlexander B Filimonov y Nikolay B Filimonov. “The peculiarities of application of the potential fields method for the problems of local navigation of mobile robots”. En: 2018XIV International Scientific-Technical Conference on Actual Problems of Electronics Instrument Engineering (APEIE). IEEE. 2018, págs. 208-211.spa
dc.relation.referencesRajat Kumar Panda y BB Choudhury. “An effective path planning of mobile robot using genetic algorithm”. En: 2015 IEEE International Conference on Computational Intelligence & Communication Technology. IEEE. 2015, págs. 287-291.spa
dc.relation.referencesMichel Aractingi y col. “Improving the Generalization of Visual Navigation Policies using Invariance Regularization”. En: (2019).spa
dc.relation.referencesXiangyun Meng y col. “Neural autonomous navigation with riemannian motion policy”. En: 2019 International Conference on Robotics and Automation (ICRA). IEEE.2019, págs. 8860-8866.spa
dc.relation.referencesRamon Gonzalez, Marius Kloetzer y Cristian Mahulea. “Comparative study of trajectories resulted from cell decomposition path planning approaches”. En: 2017 21st International Conference on System Theory, Control and Computing (ICSTCC). IEEE.2017, págs. 49-54.spa
dc.relation.referencesYuki Kato, Koji Kamiyama y Kazuyuki Morioka. “Autonomous robot navigation system with learning based on deep Q-network and topological maps”. En: 2017IEEE/SICE International Symposium on System Integration (SII). IEEE. 2017, págs. 1040-1046.spa
dc.relation.referencesColin Greatwood y Arthur G Richards. “Reinforcement learning and model predictive control for robust embedded quadrotor guidance and control”. En: Autonomous Robots 43.7 (2019), págs. 1681-1693.spa
dc.relation.referencesHaobin Shi y col. “End-to-end navigation strategy with deep reinforcement learningfor mobile robots”. En: IEEE Transactions on Industrial Informatics 16.4 (2019), págs. 2393-2402.spa
dc.relation.referencesJake Bruce y col. “One-shot reinforcement learning for robot navigation with interactive replay”. En: arXiv preprint arXiv:1711.10137(2017).spa
dc.relation.referencesCarlos Celemin, Javier Ruiz-del Solar y Jens Kober. “A fast hybrid reinforcement learning framework with human corrective feedback”. En: Autonomous Robots43.5(2019), págs. 1173-1186.spa
dc.relation.referencesSeung-Ho Han y col. “Sensor-Based Mobile Robot Navigation via Deep Reinforcement Learning”. En: 2018 IEEE International Conference on Big Data and Smart Computing (BigComp). IEEE. 2018, págs. 147-154.spa
dc.relation.referencesLiulong Ma y col. “Learning to Navigate in Indoor Environments: from Memorizing to Reasoning”. En: arXiv preprint arXiv:1904.06933(2019).spa
dc.relation.referencesDmytro Mishkin, Alexey Dosovitskiy y Vladlen Koltun. “Benchmarking classic and learned navigation in complex 3D environments”. En: arXiv preprint arXiv:1901.10915(2019).spa
dc.relation.referencesYe Zhou, Erik-Jan van Kampen y Qiping Chu. “Hybrid Hierarchical Reinforcement Learning for online guidance and navigation with partial observability”. En: Neuro-computing 331 (2019), págs. 443-457.spa
dc.relation.referencesLv Qiang y col. “A model-free mapless navigation method for mobile robot using reinforcement learning”. En: 2018 Chinese Control And Decision Conference (CCDC).IEEE. 2018, págs. 3410-3415.spa
dc.relation.referencesMihai Duguleana y Gheorghe Mogan. “Neural networks based reinforcement learning for mobile robots obstacle avoidance”. En: Expert Systems with Applications 62 (2016), págs. 104-115.spa
dc.relation.referencesFidel Aznar Gregori, Mar Pujol, Ramón Rizo y col. “Obtaining fault tolerance avoidance behavior using deep reinforcement learning”. En: (2019).spa
dc.relation.referencesT Tongloy y col. “Asynchronous deep reinforcement learning for the mobile robot navigation with supervised auxiliary tasks”. En: 2017 2nd International Conference on Robotics and Automation Engineering (ICRAE). IEEE. 2017, págs. 68-72.spa
dc.relation.referencesXingyu Zhao y col. “Asynchronous reinforcement learning algorithms for solving discrete space path planning problems”. En: Applied Intelligence 48.12 (2018), págs. 4889-4904.spa
dc.relation.referencesHongjie Geng y col. “Reinforcement extreme learning machine for mobile robot navigation”. En: Proceedings of ELM-2016. Springer, 2018, págs. 61-73.spa
dc.relation.referencesXingyu Zhao y col. “Asynchronous reinforcement learning algorithms for solving discrete space path planning problems”. En: Applied Intelligence 48.12 (2018), págs. 4889-4904.spa
dc.relation.referencesBaochang Zhang y col. “Cooperative and geometric learning for path planning of UAVs”. En: 2013 International Conference on Unmanned Aircraft Systems (ICUAS).IEEE. 2013, págs. 69-78.spa
dc.relation.referencesRushikesh Kamalapurkar y col. Reinforcement learning for optimal feedback control. Springer, 2018.spa
dc.relation.referencesChristopher Gatti. Design of experiments for reinforcement learning. Springer, 2014.spa
dc.relation.referencesTodd Hester. TEXPLORE: Temporal Difference Reinforcement Learning for Robotsand Time-Constrained Domains. Springer, 2013.spa
dc.relation.referencesWen Wu y col. “Deep Deterministic Policy Gradient with Clustered Prioritized Sampling”. En: International Conference on Neural Information Processing. Springer. 2018, págs. 645-654.spa
dc.relation.referencesShauharda Khadka y col. “Collaborative evolutionary reinforcement learning”. En: arXiv preprint arXiv:1905.00976 (2019).spa
dc.relation.referencesYoko Sasaki y col. “A3C Based Motion Learning for an Autonomous Mobile Robot in Crowds”. En: 2019 IEEE International Conference on Systems, Man and Cybernetics (SMC). IEEE. 2019, págs. 1036-1042.spa
dc.relation.referencesAlfonso B Labao, Mygel Andrei M Martija y Prospero C Naval. “A3C-GS: Adaptive Moment Gradient Sharing With Locks for Asynchronous Actor-Critic Agents”. En: IEEE Transactions on Neural Networks and Learning Systems (2020).spa
dc.relation.referencesPaul Christiano y col. “Transfer from simulation to real world through learning deep inverse dynamics model”. En: arXiv preprint arXiv:1610.03518 (2016).spa
dc.relation.referencesJoohyun Woo y Nakwan Kim. “Vision based obstacle detection and collision risk esti-mation of an unmanned surface vehicle”. En:2016 13th International Conference onUbiquitous Robots and Ambient Intelligence (URAI). IEEE. 2016, p ́ags. 461-465spa
dc.rightsDerechos reservados - Universidad Nacional de Colombiaspa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseReconocimiento 4.0 Internacionalspa
dc.rights.spaAcceso abiertospa
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/spa
dc.subject.ddc000 - Ciencias de la computación, información y obras generales::003 - Sistemasspa
dc.subject.proposalNavigationeng
dc.subject.proposalNavegaciónspa
dc.subject.proposalRobóticaspa
dc.subject.proposalRoboticseng
dc.subject.proposalReinforcement learningeng
dc.subject.proposalAprendizaje por refuerzospa
dc.subject.proposalArtificial intelligenceeng
dc.subject.proposalInteligencia artificialspa
dc.subject.proposalMachine learningeng
dc.subject.proposalAprendizaje de máquinaspa
dc.titleDiseño de una estrategia para la planeación de rutas de navegación autónoma de un robot móvil en entornos interiores usando un algoritmo de aprendizaje automáticospa
dc.title.alternativeDesign of a strategy for the planning of autonomous navigation routes of a mobile robot in indoor environments using a machine learning algorithmspa
dc.typeTrabajo de grado - Maestríaspa
dc.type.coarhttp://purl.org/coar/resource_type/c_bdccspa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/masterThesisspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
15384541.2020.pdf
Tamaño:
1.22 MB
Formato:
Adobe Portable Document Format
Descripción:
Tesis de Maestría en Ingeniería - Ingeniería de Sistemas

Bloque de licencias

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
license.txt
Tamaño:
3.8 KB
Formato:
Item-specific license agreed upon to submission
Descripción: