Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981659
Christian Dengler, B. Lohmann
In this contribution, we develop a feedback controller for a wheeled inverted pendulum in the form of a neural network that is not only stabilizing the unstable system, but also allows the wheeled robot to drive to arbitrary positions within a certain radius and take a desired orientation, without the need to compute a feasible trajectory to the desired position online. While some techniques from the reinforcement learning community can be used to optimize the parameters of a general feedback controller, i.e. policy gradient methods, the method used in this work is an approach related to imitation learning or learning from demonstration. The demonstration data however does not result from e.g. a human demonstrator, but is a set of precomputed optimal trajectories. The neural network is trained to imitate the behavior of those optimal trajectories. We show that a good choice of initial states and a large number of training targets can be used to alleviate a problem of imitation learning, namely deviating from training trajectories, and we demonstrate results in simulation as well as on the physical system.
{"title":"Neural network position and orientation control of an inverted pendulum on wheels","authors":"Christian Dengler, B. Lohmann","doi":"10.1109/ICAR46387.2019.8981659","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981659","url":null,"abstract":"In this contribution, we develop a feedback controller for a wheeled inverted pendulum in the form of a neural network that is not only stabilizing the unstable system, but also allows the wheeled robot to drive to arbitrary positions within a certain radius and take a desired orientation, without the need to compute a feasible trajectory to the desired position online. While some techniques from the reinforcement learning community can be used to optimize the parameters of a general feedback controller, i.e. policy gradient methods, the method used in this work is an approach related to imitation learning or learning from demonstration. The demonstration data however does not result from e.g. a human demonstrator, but is a set of precomputed optimal trajectories. The neural network is trained to imitate the behavior of those optimal trajectories. We show that a good choice of initial states and a large number of training targets can be used to alleviate a problem of imitation learning, namely deviating from training trajectories, and we demonstrate results in simulation as well as on the physical system.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"13 1","pages":"350-355"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89274234","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981635
Paul Glogowski, Kai Lemmerz, A. Hypki, B. Kuhlenkötter
Speed and separation monitoring (SSM) is one of the four permitted collaborative operations in human-robot interaction (HRI). Current standards and guidelines provide users and system integrators with a simple basis to calculate permissible separation distances between human workers and robots. However, high impact factors due to various simplifications result in oversized safety zones, which in practice often leads to difficulties in layout and process design. The very large safety zones that have been required so far are one of the existing obstacles to the implementation of HRI applications, especially in SSM. This paper describes extension approaches to determine the dynamic separation distance more precisely and to calculate the adapted robot speed. The developed methods are integrated into an existing HRI simulation tool based on the Robot Operating System (ROS) and finally analyzed. Taking into account the normative conditions, the implemented methods enable users and system integrators to simulate, analyze and optimize HRI scenarios already in the planning phase.
{"title":"Extended Calculation of the Dynamic Separation Distance for Robot Speed Adaption in the Human-Robot Interaction","authors":"Paul Glogowski, Kai Lemmerz, A. Hypki, B. Kuhlenkötter","doi":"10.1109/ICAR46387.2019.8981635","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981635","url":null,"abstract":"Speed and separation monitoring (SSM) is one of the four permitted collaborative operations in human-robot interaction (HRI). Current standards and guidelines provide users and system integrators with a simple basis to calculate permissible separation distances between human workers and robots. However, high impact factors due to various simplifications result in oversized safety zones, which in practice often leads to difficulties in layout and process design. The very large safety zones that have been required so far are one of the existing obstacles to the implementation of HRI applications, especially in SSM. This paper describes extension approaches to determine the dynamic separation distance more precisely and to calculate the adapted robot speed. The developed methods are integrated into an existing HRI simulation tool based on the Robot Operating System (ROS) and finally analyzed. Taking into account the normative conditions, the implemented methods enable users and system integrators to simulate, analyze and optimize HRI scenarios already in the planning phase.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"8 1","pages":"205-212"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78163753","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981561
G. Garcia, Filipe A. S. Rocha, M. Torre, W. Serrantola, F. Lizarralde, Andre Franca, G. Pessin, G. Freitas
Belt conveyors play an essential role in the transportation of dry bulk material in different industries. Inspecting conveyor belts structures and its components, such as idlers rolls, is a fundamental task to guarantee the proper production flow. Traditionally, these are cognitive inspections based on sound and vision. In this paper we describe a novel procedure to inspect belt conveyor structures with a robotic device. The system is composed by (i) a mobile platform capable of moving in different terrains, overcoming obstacles and traversing stairs with different slopes, (ii) a robotic manipulator with six degrees of freedom, and (iii) a set of sensors including microphone, accelerometers, laser, and cameras. Preliminary field tests validated the proposed system for mining operations allowing the identification of enhancements regarding platform mobility and control strategy. Based on the kinematic model, we present a control method to command both the mobile platform and robotic manipulator considering the robotic device as a whole-body system. The strategy is validated through simulations using ROS and V-REP.
{"title":"ROSI: A Novel Robotic Method for Belt Conveyor Structures Inspection","authors":"G. Garcia, Filipe A. S. Rocha, M. Torre, W. Serrantola, F. Lizarralde, Andre Franca, G. Pessin, G. Freitas","doi":"10.1109/ICAR46387.2019.8981561","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981561","url":null,"abstract":"Belt conveyors play an essential role in the transportation of dry bulk material in different industries. Inspecting conveyor belts structures and its components, such as idlers rolls, is a fundamental task to guarantee the proper production flow. Traditionally, these are cognitive inspections based on sound and vision. In this paper we describe a novel procedure to inspect belt conveyor structures with a robotic device. The system is composed by (i) a mobile platform capable of moving in different terrains, overcoming obstacles and traversing stairs with different slopes, (ii) a robotic manipulator with six degrees of freedom, and (iii) a set of sensors including microphone, accelerometers, laser, and cameras. Preliminary field tests validated the proposed system for mining operations allowing the identification of enhancements regarding platform mobility and control strategy. Based on the kinematic model, we present a control method to command both the mobile platform and robotic manipulator considering the robotic device as a whole-body system. The strategy is validated through simulations using ROS and V-REP.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"3 1","pages":"326-331"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78479542","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981556
Iuro B. P. Nascimento, A. Ferramosca, L. Pimenta, G. Raffo
This work presents a Nonlinear Model Predictive Control strategy for a quadrotor UAV with obstacle avoidance capability in a 3D unknown environment with static obstacles. The system aims to reach the target in minimum time while avoiding obstacles and also to take into account the energy of states and inputs. Sensor information is processed to detect the obstacles and obtain the inequality constraints of an obstacle-free zone. Numerical results are presented to attest the performance of the system.
{"title":"NMPC Strategy for a Quadrotor UAV in a 3D Unknown Environment","authors":"Iuro B. P. Nascimento, A. Ferramosca, L. Pimenta, G. Raffo","doi":"10.1109/ICAR46387.2019.8981556","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981556","url":null,"abstract":"This work presents a Nonlinear Model Predictive Control strategy for a quadrotor UAV with obstacle avoidance capability in a 3D unknown environment with static obstacles. The system aims to reach the target in minimum time while avoiding obstacles and also to take into account the energy of states and inputs. Sensor information is processed to detect the obstacles and obtain the inequality constraints of an obstacle-free zone. Numerical results are presented to attest the performance of the system.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"67 1","pages":"179-184"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73769006","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981617
J. C. V. Soares, M. Gattass, M. Meggiolaro
Simultaneous Localization and Mapping (SLAM) is a fundamental problem in mobile robotics. However, the majority of Visual SLAM algorithms assume a static scenario, limiting their applicability in real-world environments. Dealing with dynamic content in Visual SLAM is still an open problem, with solutions usually relying on direct or feature-based methods. Deep learning techniques can improve the SLAM solution in environments with a priori dynamic objects, providing high-level information of the scene. This paper presents a new approach to SLAM in human populated environments using deep learning-based techniques. The system is built on ORB-SLAM2, a state-of-the-art SLAM system. The proposed methodology is evaluated using a benchmark dataset, outperforming other Visual SLAM methods in highly dynamic scenarios.
{"title":"Visual SLAM in Human Populated Environments: Exploring the Trade-off between Accuracy and Speed of YOLO and Mask R-CNN","authors":"J. C. V. Soares, M. Gattass, M. Meggiolaro","doi":"10.1109/ICAR46387.2019.8981617","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981617","url":null,"abstract":"Simultaneous Localization and Mapping (SLAM) is a fundamental problem in mobile robotics. However, the majority of Visual SLAM algorithms assume a static scenario, limiting their applicability in real-world environments. Dealing with dynamic content in Visual SLAM is still an open problem, with solutions usually relying on direct or feature-based methods. Deep learning techniques can improve the SLAM solution in environments with a priori dynamic objects, providing high-level information of the scene. This paper presents a new approach to SLAM in human populated environments using deep learning-based techniques. The system is built on ORB-SLAM2, a state-of-the-art SLAM system. The proposed methodology is evaluated using a benchmark dataset, outperforming other Visual SLAM methods in highly dynamic scenarios.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"8 1","pages":"135-140"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75364881","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981597
D. Dornellas, F. Rosa, A. Bernardino, R. Ribeiro, J. Santos-Victor
This work tested the applicability of a navigation system based on cameras and inertial sensors for a quadrotor UAV, motivated by the desire to expand its operational envelope to regions with low GPS signal reception. A standalone navigation module composed of a stereo camera pair, an IMU, synchronization electronics and a SoC computational board was developed and provided with OKVIS, a recent open-source VIO (Visual-Inertial Odometry) algorithm. In order to quantitatively assess its performance, an indoors dataset was recorded in a controlled environment, with precise ground-truth from a motion capture system. The system was integrated with a quadrotor UAV, functioning as an additional GPS sensor from the perspective of the onboard autopilot. For this end, the VIO trajectory data was georeferenced using information from the onboard GPS receiver, as well as from the orientation estimator embedded in the IMU. To the best of our knowledge, our system is the first to follow this integration approach, this being one of the main contributions of this work. Having validated the system with handheld testing, flight tests were performed. We show, qualitatively, that our system effectively yields improved trajectory estimates under low GPS signal reception.
{"title":"GPS emulation via visual-inertial odometry for inspection drones","authors":"D. Dornellas, F. Rosa, A. Bernardino, R. Ribeiro, J. Santos-Victor","doi":"10.1109/ICAR46387.2019.8981597","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981597","url":null,"abstract":"This work tested the applicability of a navigation system based on cameras and inertial sensors for a quadrotor UAV, motivated by the desire to expand its operational envelope to regions with low GPS signal reception. A standalone navigation module composed of a stereo camera pair, an IMU, synchronization electronics and a SoC computational board was developed and provided with OKVIS, a recent open-source VIO (Visual-Inertial Odometry) algorithm. In order to quantitatively assess its performance, an indoors dataset was recorded in a controlled environment, with precise ground-truth from a motion capture system. The system was integrated with a quadrotor UAV, functioning as an additional GPS sensor from the perspective of the onboard autopilot. For this end, the VIO trajectory data was georeferenced using information from the onboard GPS receiver, as well as from the orientation estimator embedded in the IMU. To the best of our knowledge, our system is the first to follow this integration approach, this being one of the main contributions of this work. Having validated the system with handheld testing, flight tests were performed. We show, qualitatively, that our system effectively yields improved trajectory estimates under low GPS signal reception.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"26 1","pages":"755-760"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73700571","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981580
Arthicha Srisuchinnawong, Dong Shao, Potiwat Ngamkajornwiwat, Pitiwut Teerakittikul, Z. Dai, A. Ji, P. Manoonpong
Geckos are highly adaptable creatures, able to scale a variety of slopes, including walls, and can change their gait depending on their environment. Roboticists have tried to implement this behaviour in gecko robots. So far, an open-loop controlled robot without a tail that uses only one specific gait can climb to a 50° slope. In this paper, we propose neural control that allows a gecko robot to climb to a 70° slope by generating different gaits for various slope angles. The control consists of three main components: a central pattern generator (CPG) for generating various rhythmic patterns, CPG post-processing for shaping the CPG signals, and a delay line for transmitting the shaped CPG signals to drive the legs of the gecko robot. The robot uses a body inclination sensor to provide sensory feedback for gait adaptation. When the incline is below 35°, the robot walks with a predefined fast trot gait. If the incline is increased, it will change its gait from the trot gait to an intermediate gait, followed by a slow wave gait, which is both the most stable and the slowest gait, for climbing the steepest slopes. Using this walking strategy, the robot can efficiently climb a variety of slopes using different gaits and can automatically adapt its gait to maximise speed while ensuring stability.
{"title":"Neural Control for Gait Generation and Adaptation of a Gecko Robot","authors":"Arthicha Srisuchinnawong, Dong Shao, Potiwat Ngamkajornwiwat, Pitiwut Teerakittikul, Z. Dai, A. Ji, P. Manoonpong","doi":"10.1109/ICAR46387.2019.8981580","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981580","url":null,"abstract":"Geckos are highly adaptable creatures, able to scale a variety of slopes, including walls, and can change their gait depending on their environment. Roboticists have tried to implement this behaviour in gecko robots. So far, an open-loop controlled robot without a tail that uses only one specific gait can climb to a 50° slope. In this paper, we propose neural control that allows a gecko robot to climb to a 70° slope by generating different gaits for various slope angles. The control consists of three main components: a central pattern generator (CPG) for generating various rhythmic patterns, CPG post-processing for shaping the CPG signals, and a delay line for transmitting the shaped CPG signals to drive the legs of the gecko robot. The robot uses a body inclination sensor to provide sensory feedback for gait adaptation. When the incline is below 35°, the robot walks with a predefined fast trot gait. If the incline is increased, it will change its gait from the trot gait to an intermediate gait, followed by a slow wave gait, which is both the most stable and the slowest gait, for climbing the steepest slopes. Using this walking strategy, the robot can efficiently climb a variety of slopes using different gaits and can automatically adapt its gait to maximise speed while ensuring stability.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"36 1","pages":"468-473"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73870262","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981560
D. A. Lima, G. Oliveira
Collective intelligence has attracted attention of many researchers seeking to understand different real-world problems. In swarm robotics, the study of this area has revolutionized control algorithms, especially when they are aligned with other techniques that allow the easy programming of these robotic equipment. This work proposes a control algorithm for homogeneous and heterogeneous robots teams that perform garbage collection task based on cellular automata ants and Tabu search. Unlike precursor methods, in this work both searching and homing states are stochastic and the deposition and decline pheromone parameters are dynamic over time. From simulations it was possible to show that the new controller is adaptable to different parameters and at the same time is efficient in the garbage collection task for swarm robotics.
{"title":"Stochastic Cellular Automata Ant Memory model for swarm robots performing efficiently the garbage collection task","authors":"D. A. Lima, G. Oliveira","doi":"10.1109/ICAR46387.2019.8981560","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981560","url":null,"abstract":"Collective intelligence has attracted attention of many researchers seeking to understand different real-world problems. In swarm robotics, the study of this area has revolutionized control algorithms, especially when they are aligned with other techniques that allow the easy programming of these robotic equipment. This work proposes a control algorithm for homogeneous and heterogeneous robots teams that perform garbage collection task based on cellular automata ants and Tabu search. Unlike precursor methods, in this work both searching and homing states are stochastic and the deposition and decline pheromone parameters are dynamic over time. From simulations it was possible to show that the new controller is adaptable to different parameters and at the same time is efficient in the garbage collection task for swarm robotics.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"14 1","pages":"708-713"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74548087","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981551
Mohammadreza Yavari, K. Gupta, M. Mehrandezh
Kinodynamic-RRT* provides a sampling based asymptotically-optimal solution for motion planning of kinematically- and dynamically-constrained robots. For nonlinear systems, normally, the time- and energy-clamped steering function solutions needed within the RRT* use numerical iterative schemes such as shooting methods, which are computationally cumbersome. The number of calls to these solvers increases with the size of the tree. Hence, the time complexity of finding feasible steering functions prevents kinodynamic-RRT* for non-linear systems from being utilized in realtime or in situations where fast planning and re-planning are needed. Kinematic/dynamic constraints reduction to make the steering functions solvable in real time has been proposed in literature, however, these methods would affect the optimality of the solution. In this paper, we propose a lazy-steering kinodynmaic RRT* in which, machine learning techniques are used to (1) predict if a randomly-selected node is steerable to, and (2) if the steering is deemed feasible, what would be the estimated energy cost associated, when executing it. This provides a promising framework for implementing Kinodynamic-RRT* in which the use of numerical methods is delayed (hence the name lazy steering) until a potential collision free path has been found, and only then the numerical techniques are invoked. This results in a huge improvement in the run time with little trade off on optimality. Our proposed method was tested via simulation for motion planning of an under-actuated, non-holonomic, quadcopter with nonlinear dynamics in an environment cluttered with obstacles. The lazy-steering RRT* was faster than its counterpart (which was based on some recent works) by two orders of magnitude.
{"title":"Lazy Steering RRT*: An Optimal Constrained Kinodynamic Neural Network Based Planner with no In-Exploration Steering","authors":"Mohammadreza Yavari, K. Gupta, M. Mehrandezh","doi":"10.1109/ICAR46387.2019.8981551","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981551","url":null,"abstract":"Kinodynamic-RRT* provides a sampling based asymptotically-optimal solution for motion planning of kinematically- and dynamically-constrained robots. For nonlinear systems, normally, the time- and energy-clamped steering function solutions needed within the RRT* use numerical iterative schemes such as shooting methods, which are computationally cumbersome. The number of calls to these solvers increases with the size of the tree. Hence, the time complexity of finding feasible steering functions prevents kinodynamic-RRT* for non-linear systems from being utilized in realtime or in situations where fast planning and re-planning are needed. Kinematic/dynamic constraints reduction to make the steering functions solvable in real time has been proposed in literature, however, these methods would affect the optimality of the solution. In this paper, we propose a lazy-steering kinodynmaic RRT* in which, machine learning techniques are used to (1) predict if a randomly-selected node is steerable to, and (2) if the steering is deemed feasible, what would be the estimated energy cost associated, when executing it. This provides a promising framework for implementing Kinodynamic-RRT* in which the use of numerical methods is delayed (hence the name lazy steering) until a potential collision free path has been found, and only then the numerical techniques are invoked. This results in a huge improvement in the run time with little trade off on optimality. Our proposed method was tested via simulation for motion planning of an under-actuated, non-holonomic, quadcopter with nonlinear dynamics in an environment cluttered with obstacles. The lazy-steering RRT* was faster than its counterpart (which was based on some recent works) by two orders of magnitude.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"3 1","pages":"400-407"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73313309","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}
Pub Date : 2019-12-01DOI: 10.1109/ICAR46387.2019.8981558
M. Fonseca, Bruno Vilhena Adorno, P. Fraisse
The ill-conditioning of the joint-space inertia matrix plays an important role in the dynamic behavior of robot manipulators, as well as in the controllers' performance. Indeed, due to the ill-conditioning, small perturbations in the system can produce large changes in the numerical solutions, which can lead to instability. Moreover, this characteristic is intrinsic to a phenomenon of ill-conditioning in the mechanism itself, which suggests that it may be more difficult to control the mechanism. In this context, this paper proposes an adaptive controller to be used together with an algorithm that ensures better conditioning of the inertia matrix. To evaluate the proposed technique, we compared it with two widely used controllers via statistical analysis. The results showed that the proposed adaptive controller presents a better performance than the one based on the inverse dynamics with feedback linearization, and similar results when compared to a PID controller with gravity compensation.
{"title":"An Adaptive Controller with Guarantee of Better Conditioning of the Robot Manipulator Joint-Space Inertia Matrix","authors":"M. Fonseca, Bruno Vilhena Adorno, P. Fraisse","doi":"10.1109/ICAR46387.2019.8981558","DOIUrl":"https://doi.org/10.1109/ICAR46387.2019.8981558","url":null,"abstract":"The ill-conditioning of the joint-space inertia matrix plays an important role in the dynamic behavior of robot manipulators, as well as in the controllers' performance. Indeed, due to the ill-conditioning, small perturbations in the system can produce large changes in the numerical solutions, which can lead to instability. Moreover, this characteristic is intrinsic to a phenomenon of ill-conditioning in the mechanism itself, which suggests that it may be more difficult to control the mechanism. In this context, this paper proposes an adaptive controller to be used together with an algorithm that ensures better conditioning of the inertia matrix. To evaluate the proposed technique, we compared it with two widely used controllers via statistical analysis. The results showed that the proposed adaptive controller presents a better performance than the one based on the inverse dynamics with feedback linearization, and similar results when compared to a PID controller with gravity compensation.","PeriodicalId":6606,"journal":{"name":"2019 19th International Conference on Advanced Robotics (ICAR)","volume":"44 1","pages":"111-116"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90407460","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：