Towards autonomous robot systems : 22nd annual conference, TAROS 2021, Lincoln, UK, September 8-10, 2021 : proceedings / / Charles Fox [and five others] editors

• Pubblicazione: Cham, Switzerland : , : Springer, , [2021]; ©2021
• Descrizione fisica: 1 online resource (451 pages)
• Disciplina: 629.892
• Soggetto topico: Autonomous robots; Artificial intelligence; Machine learning
• Persona (resp. second.): FoxCharles
• Nota di contenuto: Intro -- Preface -- Organization -- Contents -- Algorithms -- A Study on Dense and Sparse (Visual) Rewards in Robot Policy Learning -- 1 Introduction -- 2 Related Work -- 3 Research Methods -- 3.1 Problem Formulation -- 3.2 Rewards -- 3.3 Task Success Classifiers -- 3.4 Training Methodology -- 4 Experiments and Results -- 4.1 Training Tasks -- 4.2 Success Classifiers Results -- 4.3 Experimental Results of the DRL Agents -- 5 Conclusion and Future Work -- References -- An Open-Source Multi-goal Reinforcement Learning Environment for Robotic Manipulation with Pybullet -- 1 Introduction -- 2 Environment -- 2.1 Single-Step Tasks -- 2.2 Multi-step Tasks -- 2.3 APIs and Programming Style -- 3 Benchmark and Discussion -- 3.1 Reproducing Hindsight Experience Replay on Single-Step Tasks -- 3.2 Benchmarking Multi-step Tasks -- 3.3 Challenges and Opportunities -- 4 Conclusion -- References -- CPG-Actor: Reinforcement Learning for Central Pattern Generators -- 1 Introduction -- 1.1 Related Work -- 1.2 Contributions -- 2 Methodology -- 2.1 Differentiable Central Pattern Generators -- 3 Evaluation -- 3.1 Experimental Setup -- 4 Results -- 4.1 CPG-Actor and Previous Baselines, Comparison -- 4.2 Evaluation of Progressive Task Achievement -- 5 Discussion and Future Work -- References -- Deep Semantic Segmentation of 3D Plant Point Clouds -- 1 Introduction -- 2 Related Work -- 2.1 3D Plant Segmentation -- 2.2 Deep Learning for 3D Plant Segmentation -- 2.3 PointNet++ -- 3 Methodology -- 3.1 Data Set -- 3.2 Network Architecture -- 3.3 Performance Metrics -- 3.4 Network Training -- 4 Evaluation -- 4.1 Quantitative Performance Evaluation -- 4.2 Qualitative Performance Evaluation -- 5 Conclusions and Future Work -- References -- Discovering Stable Robot Grasps for Unknown Objects in Presence of Uncertainty Using Bayesian Models -- 1 Introduction -- 2 Related Work. ; 3 Methodology -- 3.1 Object Extraction from Point Cloud -- 3.2 Grasp Metric Calculation -- 3.3 Probabilistic Modelling -- 4 Implementation -- 4.1 Configuration -- 4.2 Protocol -- 5 Results -- 6 Conclusion -- References -- Improving SLAM in Pipe Networks by Leveraging Cylindrical Regularity -- 1 Introduction -- 2 Related Work -- 3 System Overview -- 4 Cylinder Detection -- 4.1 Cylinder Representation and Estimation -- 4.2 Cylinder Detection -- 5 Bundle Adjustment with Cylindrical Regularity -- 6 Performance Validation and Evaluation -- 6.1 Synthetic Data -- 6.2 Real Data -- 6.3 Discussion -- 7 Conclusion -- References -- CRH*: A Deadlock Free Framework for Scalable Prioritised Path Planning in Multi-robot Systems -- 1 Introduction -- 2 Related Works -- 2.1 Heuristics -- 2.2 Rescheduling -- 2.3 Path Finding -- 3 Design and Development -- 3.1 Overview -- 3.2 A* Adaptations -- 3.3 Framework Facilities -- 4 Experiments -- 4.1 Evaluation of CRH* -- 4.2 Evaluation of Optimisation Improvements -- 5 Conclusion -- References -- Task-Based Ad-hoc Teamwork with Adversary -- 1 Introduction -- 2 Related Works -- 3 Methodology -- 3.1 Ad-Hoc Teamwork with Adversaries -- 3.2 On-Line Estimators for Ad-hoc Task Allocation with Adversary -- 3.3 OEATA-A Fundamentals -- 3.4 Sets of Estimators -- 3.5 Process of Estimation -- 4 Experiments -- 4.1 Level-Based Foraging Domain -- 4.2 Results -- 5 Conclusion -- References -- Human-Robot Cooperative Lifting Using IMUs and Human Gestures -- 1 Introduction -- 2 Human-Robot Cooperative Lifting Using IMUs and Gestures -- 2.1 Posture and Gesture Estimation -- 2.2 Cooperation Roles and States in Cooperative Lifting -- 2.3 Human-Robot Cooperative Lifting of a Table -- 3 Experimental Setup and Results -- 3.1 Setup -- 3.2 Calibration -- 3.3 Results -- 4 Discussion and Future Work -- 5 Conclusions -- References. ; Reinforcement Learning-Based Mapless Navigation with Fail-Safe Localisation -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 System Description -- 3.2 Localisation Algorithm -- 3.3 Reinforcement Learning Agent -- 4 Experiments and Results -- 4.1 Experiment Setup -- 4.2 Results -- 5 Conclusion -- References -- Collaborative Coverage for a Network of Vacuum Cleaner Robots -- 1 Introduction -- 2 Method -- 3 Results -- 4 Conclusion -- References -- Network-Aware Genetic Algorithms for the Coordination of MALE UAV Networks -- 1 Introduction -- 2 Problem Description -- 3 Power-Aware Genetic Algorithms -- 4 Network-Aware Genetic Algorithms -- 4.1 Integration with a Network Simulator -- 4.2 Network Topology Evaluation Model -- 5 Experimental Methodology -- 6 Results -- 7 Conclusions -- References -- Self-organised Flocking of Robotic Swarm in Cluttered Environments -- 1 Introduction -- 2 Flocking Method -- 2.1 Active Elastic Sheet Model -- 2.2 Extended AES Model -- 2.3 Metrics -- 2.4 Experimental Setup -- 3 Results and Discussion -- 4 Conclusion -- References -- Evaluating Feedback Modalities in a Mobile Robot for Telecare -- 1 Introduction -- 2 Materials and Methods -- 2.1 The Experimental System -- 2.2 Task -- 2.3 Research Hypotheses -- 2.4 Experimental Design -- 2.5 Dependent Measures -- 2.6 Participants -- 2.7 Procedure -- 2.8 Analysis -- 3 Results -- 3.1 Efficiency -- 3.2 Understanding -- 3.3 Effectiveness -- 3.4 User Perception -- 4 Discussion -- 4.1 Impact of Feedback Modality -- 4.2 Impact of Secondary Task Location -- 5 Conclusions and Future Work -- References -- Demonstrating the Differential Impact of Flock Heterogeneity on Multi-agent Herding -- 1 Introduction -- 1.1 Motivation -- 1.2 Related Work -- 2 Model -- 2.1 Experimental Setup -- 2.2 Sheep Agent Model -- 2.3 Dog Agent Model -- 3 Results -- 4 Conclusions -- References. ; Evaluation of an OpenCV Implementation of Structure from Motion on Open Source Data -- 1 Introduction -- 2 Algorithmic Description -- 3 Implementation and Experimental Results -- 4 Comparison -- 5 Conclusions -- References -- Benchmark of Visual and 3D Lidar SLAM Systems in Simulation Environment for Vineyards -- 1 Introduction -- 2 Related Work -- 3 SLAM Algorithms -- 3.1 RTAB-Map -- 3.2 ORB-SLAM2 -- 3.3 LIO-SAM -- 3.4 StaticMapping -- 4 Evaluation -- 4.1 Environment -- 4.2 Testing Scenarios -- 4.3 SLAM Algorithms Configurations -- 4.4 Metrics -- 4.5 Results -- 5 Conclusion -- References -- Lidar-Only Localization with 3D Pose-Feature Map -- 1 Introduction -- 2 Related Work -- 3 Framework Architecture and Background -- 3.1 Framework Architecture -- 3.2 Lidar-Only Odometry Background -- 4 Pose-Feature Map Localization -- 4.1 Pose-Feature Map -- 4.2 Adaptive Local-Map Matching -- 4.3 Integrating Localization -- 5 Experiments -- 5.1 Hardware, Datasets and Parameter Setting -- 5.2 Localization Performance -- 6 Conclusion -- References -- Toward Robust Visual Odometry Using Prior 2D Map Information and Multiple Hypothesis Particle Filtering -- 1 Introduction -- 2 Methods -- 3 Results -- 4 Conclusions -- References -- Comparison of Concentrated and Distributed Compliant Elements in a 3D Printed Gripper -- 1 Introduction -- 2 Compliant Gripper Simulation -- 3 3D Printed Grippers -- 4 Conclusion -- References -- Perception of a Humanoid Robot as an Interface for Auditory Testing -- 1 Introduction -- 2 Experimental Design -- 2.1 Voice Cue Sensitivity -- 2.2 Gender Categorization -- 2.3 Emotion Identification -- 2.4 Speech-on-Speech Perception -- 2.5 Video Analysis -- 3 Further Work -- References -- Deep Learning Traversability Estimator for Mobile Robots in Unstructured Environments -- 1 Introduction -- 2 Traversability Prediction Model -- 2.1 Input Features. ; 2.2 Network Architecture -- 2.3 Robot Model and Failure Events -- 3 Dataset Generation -- 3.1 OpenSimplex Synthetic Maps Generation -- 3.2 Dataset Collection and Training -- 4 Results -- 4.1 Prediction Performance - Synthetic Dataset -- 4.2 Prediction Performance - Planetary Mission Use Case -- 5 Conclusion and Future Work -- References -- Systems -- Predicting Artist Drawing Activity via Multi-camera Inputs for Co-creative Drawing -- 1 Introduction -- 2 Background -- 3 Research Set-Up -- 4 Drawing Data Gathering Study -- 5 Visual Based Models -- 6 Experiments and Results -- 7 Discussion and Limitations -- 8 Summary and Future Work -- References -- 3D Printed Mechanically Modular Two-Degree-Of-Freedom Robotic Segment Utilizing Variable-Stiffness Actuators -- 1 Introduction -- 2 Design Considerations -- 3 Segment Design -- 4 Control Electronics -- 5 Mechanical Construction -- 6 Position Sensing and Torque Estimation -- 7 Characterizing VSA Dynamics -- 8 Control and Simulation Using ROS -- 9 MATLAB Simulations and Demo -- 10 Discussion -- References -- Design of a Multimaterial 3D-Printed Soft Actuator with Bi-directional Variable Stiffness -- 1 Introduction -- 2 Design and Fabrication of the Actuator -- 3 Control Setup -- 4 Testing -- 5 Conclusion -- References -- Designing a Multi-locomotion Modular Snake Robot -- 1 Introduction -- 2 Proposed Design -- 2.1 Locomotion -- 2.2 Housing -- 2.3 Smart Servos -- 2.4 Biomimetic Snakeskin -- 3 Experiments and Discussion -- 3.1 Prototype 1 - Snakeskin Testing -- 3.2 Prototype 2 - Housing and Smart Servos -- 4 Conclusion -- References -- Deep Robot Path Planning from Demonstrations for Breast Cancer Examination -- 1 Introduction -- 2 Methodology -- 2.1 Data-Set Acquisition -- 2.2 Deep-Model -- 3 Results -- 4 Conclusion -- References. ; Priors Inspired by Speed-Accuracy Trade-Offs for Incremental Learning of Probabilistic Movement Primitives.
• Titolo autorizzato: Towards autonomous robot systems
• ISBN: 3-030-89177-1
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 996464390103316
• Lo trovi qui: Univ. di Salerno
• Serie: Lecture notes in computer science ; ; 13054.