Pub Date : 2020-11-20DOI: 10.1109/ICRAE50850.2020.9310831
I. M. D. C. Jayasundara, A. Mudugamuwa, Han Baokun, K. Perera, Y. Amarasinghe
This paper presents the design approach and development of a novel External Pipeline Robot (EPR) named ExPiRo with the capability of moving on linear segments of cylindrical structures with variable diameters in the range 100 mm to 130 mm. The robot has a passive pipe clutching mechanism created from two parallelogram four-bar linkages. The designed robot can carry payloads up to 2.2 kg. The ExPiRo prototype demonstrated the desired ability to travel on a varying diameter pipe during testing. A control system for position controlling of the robot within the pipeline is also proposed. An ADAMS-MATLAB co-simulation is conducted to evaluate the performance of the proposed control system. The control system demonstrated significant stability in reaching different goal positions.
{"title":"Design and Development of a Novel External Pipe Crawling Robot ExPiRo","authors":"I. M. D. C. Jayasundara, A. Mudugamuwa, Han Baokun, K. Perera, Y. Amarasinghe","doi":"10.1109/ICRAE50850.2020.9310831","DOIUrl":"https://doi.org/10.1109/ICRAE50850.2020.9310831","url":null,"abstract":"This paper presents the design approach and development of a novel External Pipeline Robot (EPR) named ExPiRo with the capability of moving on linear segments of cylindrical structures with variable diameters in the range 100 mm to 130 mm. The robot has a passive pipe clutching mechanism created from two parallelogram four-bar linkages. The designed robot can carry payloads up to 2.2 kg. The ExPiRo prototype demonstrated the desired ability to travel on a varying diameter pipe during testing. A control system for position controlling of the robot within the pipeline is also proposed. An ADAMS-MATLAB co-simulation is conducted to evaluate the performance of the proposed control system. The control system demonstrated significant stability in reaching different goal positions.","PeriodicalId":296832,"journal":{"name":"2020 5th International Conference on Robotics and Automation Engineering (ICRAE)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117097675","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 : 2020-11-20DOI: 10.1109/ICRAE50850.2020.9310905
Bekarys Nurtay, Tomiris Suranshy, M. Folgheraiter
This paper presents the design and modeling of a lightweight tendon actuated robotic limb. The mechanical structure consists of a sequence of four semi-rigid segments realized in thermoplastic polyurethane material and connected through torsional springs. This allows the limb to keep a straight position without the application of forces and facilitates the control of the limb while performing flexion and extension movements. A static model is presented to predict the tension of the tendons in order to reach a defined orientation. Simulations were conducted in a V-REP Python environment to demonstrate the controllability of the limb while performing simple movements and trajectories.
{"title":"Development of a Semi-Rigid Tendon Actuated Limb for Robotics Applications","authors":"Bekarys Nurtay, Tomiris Suranshy, M. Folgheraiter","doi":"10.1109/ICRAE50850.2020.9310905","DOIUrl":"https://doi.org/10.1109/ICRAE50850.2020.9310905","url":null,"abstract":"This paper presents the design and modeling of a lightweight tendon actuated robotic limb. The mechanical structure consists of a sequence of four semi-rigid segments realized in thermoplastic polyurethane material and connected through torsional springs. This allows the limb to keep a straight position without the application of forces and facilitates the control of the limb while performing flexion and extension movements. A static model is presented to predict the tension of the tendons in order to reach a defined orientation. Simulations were conducted in a V-REP Python environment to demonstrate the controllability of the limb while performing simple movements and trajectories.","PeriodicalId":296832,"journal":{"name":"2020 5th International Conference on Robotics and Automation Engineering (ICRAE)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124046714","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 : 2020-11-20DOI: 10.1109/ICRAE50850.2020.9310864
Yi Zhan, Zihao Wang, Jiarui Xu, Guoyi Yu, F. An, Wenzheng Chi, Chao Wang
This paper proposes a fast-convergence CO-ordinate Rotation DIgital Computer (CORDIC) based Generalized Voronoi Diagram (GVD) hardware accelerator for efficient robotic path exploration. Owing to the high precision contributed by fast-convergence CORDIC, the proposed GVD hardware accelerator significantly improves the accuracy of the explored paths as compared to the baseline design. Higher precision of the exploration causes shorter trajectory of the robot, which further reduces the power consumption of the entire robot system. Therefore, our design is suitable to the battery-powered small-scale robots. FPGA implementation shows that, the proposed design operating at 12-bit fixed point achieves 54% higher precision of the explored paths and 20% lower power consumption of the robot system than the baseline design, respectively.
{"title":"Fast CORDIC based Generalized-Voronoi-Diagram Hardware Accelerator for Efficient Robotic Exploration","authors":"Yi Zhan, Zihao Wang, Jiarui Xu, Guoyi Yu, F. An, Wenzheng Chi, Chao Wang","doi":"10.1109/ICRAE50850.2020.9310864","DOIUrl":"https://doi.org/10.1109/ICRAE50850.2020.9310864","url":null,"abstract":"This paper proposes a fast-convergence CO-ordinate Rotation DIgital Computer (CORDIC) based Generalized Voronoi Diagram (GVD) hardware accelerator for efficient robotic path exploration. Owing to the high precision contributed by fast-convergence CORDIC, the proposed GVD hardware accelerator significantly improves the accuracy of the explored paths as compared to the baseline design. Higher precision of the exploration causes shorter trajectory of the robot, which further reduces the power consumption of the entire robot system. Therefore, our design is suitable to the battery-powered small-scale robots. FPGA implementation shows that, the proposed design operating at 12-bit fixed point achieves 54% higher precision of the explored paths and 20% lower power consumption of the robot system than the baseline design, respectively.","PeriodicalId":296832,"journal":{"name":"2020 5th International Conference on Robotics and Automation Engineering (ICRAE)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126286477","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 : 2020-10-21DOI: 10.1109/ICRAE50850.2020.9310861
V. Jiménez, C. Schwarzl, Szilárd Josvai
Since an automotive driving vehicle is controlled by Advanced Driver-Assistance Systems (ADAS) / Automated Driving (AD) functions, the selected sensors for the perception process become a key component of the system. Therefore, the necessity of ensuring precise data is crucial. But the correctness of the data is not the only part that has to be ensured, the limitations of the different technologies to accurately sense the reality must be checked for an error-free decision making according to the current scenario. In this context, this publication presents a comparison between two different automotive radars through our self-developed robot mobile platform called SPIDER, and how they can detect different kinds of objects in the tests carried out at the ZalaZONE proving ground.
{"title":"Radar Detection Rate Comparison through a Mobile Robot Platform at the ZalaZONE Proving Ground","authors":"V. Jiménez, C. Schwarzl, Szilárd Josvai","doi":"10.1109/ICRAE50850.2020.9310861","DOIUrl":"https://doi.org/10.1109/ICRAE50850.2020.9310861","url":null,"abstract":"Since an automotive driving vehicle is controlled by Advanced Driver-Assistance Systems (ADAS) / Automated Driving (AD) functions, the selected sensors for the perception process become a key component of the system. Therefore, the necessity of ensuring precise data is crucial. But the correctness of the data is not the only part that has to be ensured, the limitations of the different technologies to accurately sense the reality must be checked for an error-free decision making according to the current scenario. In this context, this publication presents a comparison between two different automotive radars through our self-developed robot mobile platform called SPIDER, and how they can detect different kinds of objects in the tests carried out at the ZalaZONE proving ground.","PeriodicalId":296832,"journal":{"name":"2020 5th International Conference on Robotics and Automation Engineering (ICRAE)","volume":"323 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124566140","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 : 2020-10-07DOI: 10.1109/ICRAE50850.2020.9310863
O. Speidel, Maximilian Graf, Ankita Kaushik, Thanh Phan-Huu, A. Wedel, K. Dietmayer
Efficient trajectory planning for urban intersections is currently one of the most challenging tasks for an Autonomous Vehicle (AV). Courteous behavior towards other traffic participants, the AV’s comfort and its progression in the environment are the key aspects that determine the performance of trajectory planning algorithms. To capture these aspects, we propose a novel trajectory planning framework that ensures social compliance and simultaneously optimizes the AV’s comfort subject to kinematic constraints. The framework combines a local continuous optimization approach and an efficient driver model to ensure fast behavior prediction, maneuver generation and decision making over long horizons. The proposed framework is evaluated in different scenarios to demonstrate its capabilities in terms of the resulting trajectories and runtime.
{"title":"Trajectory Planning for Automated Driving in Intersection Scenarios Using Driver Models","authors":"O. Speidel, Maximilian Graf, Ankita Kaushik, Thanh Phan-Huu, A. Wedel, K. Dietmayer","doi":"10.1109/ICRAE50850.2020.9310863","DOIUrl":"https://doi.org/10.1109/ICRAE50850.2020.9310863","url":null,"abstract":"Efficient trajectory planning for urban intersections is currently one of the most challenging tasks for an Autonomous Vehicle (AV). Courteous behavior towards other traffic participants, the AV’s comfort and its progression in the environment are the key aspects that determine the performance of trajectory planning algorithms. To capture these aspects, we propose a novel trajectory planning framework that ensures social compliance and simultaneously optimizes the AV’s comfort subject to kinematic constraints. The framework combines a local continuous optimization approach and an efficient driver model to ensure fast behavior prediction, maneuver generation and decision making over long horizons. The proposed framework is evaluated in different scenarios to demonstrate its capabilities in terms of the resulting trajectories and runtime.","PeriodicalId":296832,"journal":{"name":"2020 5th International Conference on Robotics and Automation Engineering (ICRAE)","volume":"145 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124423401","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 : 2020-08-18DOI: 10.1109/ICRAE50850.2020.9310796
Wenshuai Zhao, J. P. Queralta, Qingqing Li, Tomi Westerlund
Current research directions in deep reinforcement learning include bridging the simulation-reality gap, improving sample efficiency of experiences in distributed multi-agent reinforcement learning, together with the development of robust methods against adversarial agents in distributed learning, among many others. In this work, we are particularly interested in analyzing how multi-agent reinforcement learning can bridge the gap to reality in distributed multi-robot systems where the operation of the different robots is not necessarily homogeneous. These variations can happen due to sensing mismatches, inherent errors in terms of calibration of the mechanical joints, or simple differences in accuracy. While our results are simulation-based, we introduce the effect of sensing, calibration, and accuracy mismatches in distributed reinforcement learning with proximal policy optimization (PPO). We discuss on how both the different types of perturbances and how the number of agents experiencing those perturbances affect the collaborative learning effort. The simulations are carried out using a Kuka arm model in the Bullet physics engine. This is, to the best of our knowledge, the first work exploring the limitations of PPO in multi-robot systems when considering that different robots might be exposed to different environments where their sensors or actuators have induced errors. With the conclusions of this work, we set the initial point for future work on designing and developing methods to achieve robust reinforcement learning on the presence of real-world perturbances that might differ within a multi-robot system.
{"title":"Towards Closing the Sim-to-Real Gap in Collaborative Multi-Robot Deep Reinforcement Learning","authors":"Wenshuai Zhao, J. P. Queralta, Qingqing Li, Tomi Westerlund","doi":"10.1109/ICRAE50850.2020.9310796","DOIUrl":"https://doi.org/10.1109/ICRAE50850.2020.9310796","url":null,"abstract":"Current research directions in deep reinforcement learning include bridging the simulation-reality gap, improving sample efficiency of experiences in distributed multi-agent reinforcement learning, together with the development of robust methods against adversarial agents in distributed learning, among many others. In this work, we are particularly interested in analyzing how multi-agent reinforcement learning can bridge the gap to reality in distributed multi-robot systems where the operation of the different robots is not necessarily homogeneous. These variations can happen due to sensing mismatches, inherent errors in terms of calibration of the mechanical joints, or simple differences in accuracy. While our results are simulation-based, we introduce the effect of sensing, calibration, and accuracy mismatches in distributed reinforcement learning with proximal policy optimization (PPO). We discuss on how both the different types of perturbances and how the number of agents experiencing those perturbances affect the collaborative learning effort. The simulations are carried out using a Kuka arm model in the Bullet physics engine. This is, to the best of our knowledge, the first work exploring the limitations of PPO in multi-robot systems when considering that different robots might be exposed to different environments where their sensors or actuators have induced errors. With the conclusions of this work, we set the initial point for future work on designing and developing methods to achieve robust reinforcement learning on the presence of real-world perturbances that might differ within a multi-robot system.","PeriodicalId":296832,"journal":{"name":"2020 5th International Conference on Robotics and Automation Engineering (ICRAE)","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134171377","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：