Pub Date : 2020-08-01DOI: 10.1109/CASE48305.2020.9216792
Atsuki Kiuchi, Haiyan Wang, Qiyao Wang, Takahiro Ogura, Tazu Nomoto, Chetan Gupta, T. Matsui, Susumu Serita, Chi Zhang
Supply chain inventory optimization is essential to ensure supply chain efficiency and to increase customer satisfaction. However, it is challenging because of the inherent uncertainties and complex dynamics in real-world supply chains. Researchers and practitioners have turned to simulation-based optimization methods to solve analytically intractable multi-echelon inventory optimization problems. Whereas, simulation-based optimization methods are usually computationally expensive. An efficient optimization procedure will greatly enhance the applicability of these methods. In this paper, we propose a Bayesian optimization approach along with an agent-based supply chain simulator to solve a constrained multi-echelon inventory optimization problem that requires fewer number of interactions with the simulator. Our proposed approach is compared with the most popularly used algorithm, genetic algorithm (GA). The experimental results demonstrate that the proposed method converges to the optimal solution significantly faster than GA.
{"title":"Bayesian Optimization Algorithm with Agent-based Supply Chain Simulator for Multi-echelon Inventory Management","authors":"Atsuki Kiuchi, Haiyan Wang, Qiyao Wang, Takahiro Ogura, Tazu Nomoto, Chetan Gupta, T. Matsui, Susumu Serita, Chi Zhang","doi":"10.1109/CASE48305.2020.9216792","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9216792","url":null,"abstract":"Supply chain inventory optimization is essential to ensure supply chain efficiency and to increase customer satisfaction. However, it is challenging because of the inherent uncertainties and complex dynamics in real-world supply chains. Researchers and practitioners have turned to simulation-based optimization methods to solve analytically intractable multi-echelon inventory optimization problems. Whereas, simulation-based optimization methods are usually computationally expensive. An efficient optimization procedure will greatly enhance the applicability of these methods. In this paper, we propose a Bayesian optimization approach along with an agent-based supply chain simulator to solve a constrained multi-echelon inventory optimization problem that requires fewer number of interactions with the simulator. Our proposed approach is compared with the most popularly used algorithm, genetic algorithm (GA). The experimental results demonstrate that the proposed method converges to the optimal solution significantly faster than GA.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126910994","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9216955
Pascal Becker, Christian Eichmann, A. Rönnau, R. Dillmann
The use of industrial robots for the production of small scale manufacturing or even single pieces is rarely economical. The high investment of time and money required to teach a collision-free trajectory under consideration of all boundary conditions prevents the usage of robots until now. That is why it is still common practice in the industry for individual post-processing steps to be carried out manually, although they could be automated with today’s technical possibilities. In this paper, we present an approach on how to use existing production data (STL and G-code) to generate trajectories to automate post-processing steps. These paths can then be executed by an industrial robot, for example to post-process an additive manufactured object and remove its support structures. The object may not be damaged during the process, so all movements of the robot and its tool are checked for collisions with certain parts of the object and the environment. While material is being removed, the corresponding data structure is updated accordingly to always provide a realistic representation of the current state. This approach was evaluated by removing support structures from multiple and different-shaped objects successfully. Furthermore, we used the same approach to mill pockets in material just by changing the input data.
{"title":"Automation of Post-Processing in Additive Manufacturing with Industrial Robots","authors":"Pascal Becker, Christian Eichmann, A. Rönnau, R. Dillmann","doi":"10.1109/CASE48305.2020.9216955","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9216955","url":null,"abstract":"The use of industrial robots for the production of small scale manufacturing or even single pieces is rarely economical. The high investment of time and money required to teach a collision-free trajectory under consideration of all boundary conditions prevents the usage of robots until now. That is why it is still common practice in the industry for individual post-processing steps to be carried out manually, although they could be automated with today’s technical possibilities. In this paper, we present an approach on how to use existing production data (STL and G-code) to generate trajectories to automate post-processing steps. These paths can then be executed by an industrial robot, for example to post-process an additive manufactured object and remove its support structures. The object may not be damaged during the process, so all movements of the robot and its tool are checked for collisions with certain parts of the object and the environment. While material is being removed, the corresponding data structure is updated accordingly to always provide a realistic representation of the current state. This approach was evaluated by removing support structures from multiple and different-shaped objects successfully. Furthermore, we used the same approach to mill pockets in material just by changing the input data.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126025766","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9216738
Jiajun Liu, P. Jiang
Recently, social manufacturing emerges to help micro-and-small-scale manufacturing enterprises (MSMEs) to cope with service-oriented, personalized, diverse and dynamic market demand. In social manufacturing, MSMEs are usually peer-to-peer and geographically distributed, and they self-organize into manufacturing communities (MCs) to share their manufacturing resources, order and business benefit. Therefore, the decentralized organization and operation of MCs become an emergency. To solve this problem, a consortium block chain based decentralized mechanism is established for the organization and operation of manufacturing community in social manufacturing, and a corresponding reference framework is proposed to guide MCs to implement decentralized organization and operation. On this basis, three key enabled technologies are elaborated, which are mainly about the determination of verification dataset, the design of smart contract, and the definition of endorsement policy. It is expected that the reference framework proposed in this paper will provide a possible way for MSMEs to organize and operate MC without core enterprise in social manufacturing.
{"title":"Consortium blockchain-driven decentralized organization and operation for manufacturing community in social manufacturing","authors":"Jiajun Liu, P. Jiang","doi":"10.1109/CASE48305.2020.9216738","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9216738","url":null,"abstract":"Recently, social manufacturing emerges to help micro-and-small-scale manufacturing enterprises (MSMEs) to cope with service-oriented, personalized, diverse and dynamic market demand. In social manufacturing, MSMEs are usually peer-to-peer and geographically distributed, and they self-organize into manufacturing communities (MCs) to share their manufacturing resources, order and business benefit. Therefore, the decentralized organization and operation of MCs become an emergency. To solve this problem, a consortium block chain based decentralized mechanism is established for the organization and operation of manufacturing community in social manufacturing, and a corresponding reference framework is proposed to guide MCs to implement decentralized organization and operation. On this basis, three key enabled technologies are elaborated, which are mainly about the determination of verification dataset, the design of smart contract, and the definition of endorsement policy. It is expected that the reference framework proposed in this paper will provide a possible way for MSMEs to organize and operate MC without core enterprise in social manufacturing.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"54 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128137214","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9217005
Adriano M. C. Rezende, G. P. C. Júnior, R. Fernandes, Victor R. F. Miranda, Héctor Azpúrua, G. Pessin, G. Freitas
Localization is critical for autonomous robot operation, and selecting a suitable method for pose estimation is still a challenging task. In this sense, this paper investigates different localization and navigation control strategies deployed into the EspeleoRobô, a robotic platform designed by the Brazilian mining company Vale S.A. to inspect confined areas. We compare the pose estimation algorithms based on wheel, visual and LiDAR odometry, and also Ultra-Wideband radio signals, all fused with IMU (Inertial Measurement Unit) data. Our experiments consider both teleoperated and autonomous robot operation. The robot’s autonomous navigation is based on an artificial vector fields controller, which uses the different pose estimations as feedback to guide the robot through pre-defined paths. Real experiments performed in indoor environments illustrate the performance of each estimator. Finally, preliminary mapping results states for the LiDAR SLAM (Simultaneous Localization and Mapping) approach as a promising option for practical field operations.
{"title":"Indoor Localization and Navigation Control Strategies for a Mobile Robot Designed to Inspect Confined Environments","authors":"Adriano M. C. Rezende, G. P. C. Júnior, R. Fernandes, Victor R. F. Miranda, Héctor Azpúrua, G. Pessin, G. Freitas","doi":"10.1109/CASE48305.2020.9217005","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9217005","url":null,"abstract":"Localization is critical for autonomous robot operation, and selecting a suitable method for pose estimation is still a challenging task. In this sense, this paper investigates different localization and navigation control strategies deployed into the EspeleoRobô, a robotic platform designed by the Brazilian mining company Vale S.A. to inspect confined areas. We compare the pose estimation algorithms based on wheel, visual and LiDAR odometry, and also Ultra-Wideband radio signals, all fused with IMU (Inertial Measurement Unit) data. Our experiments consider both teleoperated and autonomous robot operation. The robot’s autonomous navigation is based on an artificial vector fields controller, which uses the different pose estimations as feedback to guide the robot through pre-defined paths. Real experiments performed in indoor environments illustrate the performance of each estimator. Finally, preliminary mapping results states for the LiDAR SLAM (Simultaneous Localization and Mapping) approach as a promising option for practical field operations.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126705062","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9216808
Jian Ruan, Houde Liu, Anshun Xue, Xueqian Wang, Bin Liang
The problem of grasping is essential in robotics, in which robotic hands grasp target objects by deploying contact force. The contact between robotic hand and target object is always reduced to point-to-point while generally the actual type should be surface-to-surface. In this paper, we propose a novel surface contact model to parameterize the contact area. It computes the grasp quality of the contacts between the robotic hand and target object, then determines whether or not the surface-contact-based end effectors can resist an external wrench on target object. The key idea of the proposed approach is that the contact modeling is based on surface-to-surface contact rather than equivalent contact points which is more in line with the actual situations. Then we propose a grasp quality evaluation network based on surface-to-surface contact to evaluate grasp motion, which can capture the geometric feature of the actual contact area and classify the quality level of input grasp. Experimental results and comparisons with state-of-the-art methods (e.g. GPD and PointNetGPD) demonstrate that our approach could achieve superior performance.
{"title":"Grasp Quality Evaluation Network for Surface-to-Surface Contacts in Point Clouds","authors":"Jian Ruan, Houde Liu, Anshun Xue, Xueqian Wang, Bin Liang","doi":"10.1109/CASE48305.2020.9216808","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9216808","url":null,"abstract":"The problem of grasping is essential in robotics, in which robotic hands grasp target objects by deploying contact force. The contact between robotic hand and target object is always reduced to point-to-point while generally the actual type should be surface-to-surface. In this paper, we propose a novel surface contact model to parameterize the contact area. It computes the grasp quality of the contacts between the robotic hand and target object, then determines whether or not the surface-contact-based end effectors can resist an external wrench on target object. The key idea of the proposed approach is that the contact modeling is based on surface-to-surface contact rather than equivalent contact points which is more in line with the actual situations. Then we propose a grasp quality evaluation network based on surface-to-surface contact to evaluate grasp motion, which can capture the geometric feature of the actual contact area and classify the quality level of input grasp. Experimental results and comparisons with state-of-the-art methods (e.g. GPD and PointNetGPD) demonstrate that our approach could achieve superior performance.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125122179","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9216848
Zichun Wang, Gaowei Xu, Jingwei Wang, Min Liu, Yumin Ma
Intelligent fault diagnosis methods based on deep learning have been widely used in intelligent manufacturing. Most of these methods focus on the diagnosis of fault data with the same distribution in a single domain, but pay poor attention to the diagnosis of cross-domain fault data with different distributions. To address this problem, this paper firstly integrates the fault datasets from eight universities into a cross-domain dataset. A new model named one-dimensional improved LeNet-5 (ID ILeNet-5) is proposed for cross-domain fault diagnosis. One-dimensional convolutional operation is used for feature extraction and batch normalization technique is introduced to accelerate the network convergence in this model. The effectiveness and generalization performance of this method are verified using the aforementioned cross-domain dataset. The results demonstrate that our method outperforms the state-of-the-art transfer learning model with fewer parameters and shorter training time.
{"title":"Cross-Domain Fault Diagnosis with One-Dimensional Convolutional Neural Network*","authors":"Zichun Wang, Gaowei Xu, Jingwei Wang, Min Liu, Yumin Ma","doi":"10.1109/CASE48305.2020.9216848","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9216848","url":null,"abstract":"Intelligent fault diagnosis methods based on deep learning have been widely used in intelligent manufacturing. Most of these methods focus on the diagnosis of fault data with the same distribution in a single domain, but pay poor attention to the diagnosis of cross-domain fault data with different distributions. To address this problem, this paper firstly integrates the fault datasets from eight universities into a cross-domain dataset. A new model named one-dimensional improved LeNet-5 (ID ILeNet-5) is proposed for cross-domain fault diagnosis. One-dimensional convolutional operation is used for feature extraction and batch normalization technique is introduced to accelerate the network convergence in this model. The effectiveness and generalization performance of this method are verified using the aforementioned cross-domain dataset. The results demonstrate that our method outperforms the state-of-the-art transfer learning model with fewer parameters and shorter training time.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"174 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122049560","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9216737
Ming Wang, Peng Zhang, Peng Zheng, Junjie He, Jie Zhang, J. Bao
Dynamic disturbances such as rush job arrivals and process delay are inevitable occurrences in production environment. Dynamic job shop scheduling problem (DJSSP) is known as NP-hard combinatorial optimization problem, this paper introduces an efficient strategy for the problem. Inspired by rolling horizon strategy, the hybrid periodic and event-driven rolling horizon strategy (HRS) is presented to trigger rescheduling in a dynamic environment with process delay and rush job arrivals. Within the framework, an improved genetic algorithm (IGA) with local search is proposed to generate the rescheduling scheme of unprocessed and new jobs. To evaluate the performance of proposed algorithm, various benchmark problems and different dynamic disturbances are considered to carry out detailed experiments. The results indicate that the proposed algorithm produces superior solutions for benchmark problems and solves the DJSSP effectively with different disturbances under dynamic manufacturing environment.
{"title":"An Improved Genetic Algorithm with Local Search for Dynamic Job Shop Scheduling Problem","authors":"Ming Wang, Peng Zhang, Peng Zheng, Junjie He, Jie Zhang, J. Bao","doi":"10.1109/CASE48305.2020.9216737","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9216737","url":null,"abstract":"Dynamic disturbances such as rush job arrivals and process delay are inevitable occurrences in production environment. Dynamic job shop scheduling problem (DJSSP) is known as NP-hard combinatorial optimization problem, this paper introduces an efficient strategy for the problem. Inspired by rolling horizon strategy, the hybrid periodic and event-driven rolling horizon strategy (HRS) is presented to trigger rescheduling in a dynamic environment with process delay and rush job arrivals. Within the framework, an improved genetic algorithm (IGA) with local search is proposed to generate the rescheduling scheme of unprocessed and new jobs. To evaluate the performance of proposed algorithm, various benchmark problems and different dynamic disturbances are considered to carry out detailed experiments. The results indicate that the proposed algorithm produces superior solutions for benchmark problems and solves the DJSSP effectively with different disturbances under dynamic manufacturing environment.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122147836","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9216948
Deyuan Chen, Zhiqiang Yang, Lars Svensson, Lei Feng
An articulated vehicle is a two-body design capable of precise maneuvering around obstacles, while carrying heavy loads over rough terrain. In the context of path planning for automated articulated vehicles, it is desirable to fully utilize the maneuverability of the vehicle to enable autonomous operation in confined areas. In this paper we study the impact of model accuracy in an optimization based path planner for an articulated vehicle. For this purpose, we compare the traditional kinematic bicycle model with a two-body articulated model. We evaluate performance in terms of path length, path quality, success rate and computation time through performing test queries in artificial environments and through experiments on a full scale articulated hauler. Results show that for simple, unidirectional maneuvers, performance differences are small, but for more difficult bidirectional maneuvers, the articulated model produces shorter and higher quality paths at a higher success rate. However, the articulated model has 2.75 times longer computation time on average.
{"title":"Optimization based path planning for a two-body articulated vehicle","authors":"Deyuan Chen, Zhiqiang Yang, Lars Svensson, Lei Feng","doi":"10.1109/CASE48305.2020.9216948","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9216948","url":null,"abstract":"An articulated vehicle is a two-body design capable of precise maneuvering around obstacles, while carrying heavy loads over rough terrain. In the context of path planning for automated articulated vehicles, it is desirable to fully utilize the maneuverability of the vehicle to enable autonomous operation in confined areas. In this paper we study the impact of model accuracy in an optimization based path planner for an articulated vehicle. For this purpose, we compare the traditional kinematic bicycle model with a two-body articulated model. We evaluate performance in terms of path length, path quality, success rate and computation time through performing test queries in artificial environments and through experiments on a full scale articulated hauler. Results show that for simple, unidirectional maneuvers, performance differences are small, but for more difficult bidirectional maneuvers, the articulated model produces shorter and higher quality paths at a higher success rate. However, the articulated model has 2.75 times longer computation time on average.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131832552","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9217037
J. Karlsson, Jana Tumova
We present an approach for intention-aware motion planning in an autonomous driving scenario, where a vehicle aims to traverse a road segment as quickly as possible, while constrained by road rules encoded in syntactically co-safe linear temporal logic. We show that by combining the RRTx algorithm with trajectory prediction using Mixed Observable Markov Decision Processes (MOMDP), we can achieve least-violating behavior wrt. mission completion time and the road rules, while ensuring that the likelihood of collisions remains below a user specified threshold. We illustrate the validity of our approach using simulations of a variety of traffic scenarios.
{"title":"Intention-aware motion planning with road rules","authors":"J. Karlsson, Jana Tumova","doi":"10.1109/CASE48305.2020.9217037","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9217037","url":null,"abstract":"We present an approach for intention-aware motion planning in an autonomous driving scenario, where a vehicle aims to traverse a road segment as quickly as possible, while constrained by road rules encoded in syntactically co-safe linear temporal logic. We show that by combining the RRTx algorithm with trajectory prediction using Mixed Observable Markov Decision Processes (MOMDP), we can achieve least-violating behavior wrt. mission completion time and the road rules, while ensuring that the likelihood of collisions remains below a user specified threshold. We illustrate the validity of our approach using simulations of a variety of traffic scenarios.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114945483","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 : 2020-08-01DOI: 10.1109/CASE48305.2020.9217040
Wei-feng Lu, Zhe Hu, Jia Pan
Due to the difficulty of modeling human limb, it is very challenging to design the controller for human-robot collaboration. In this paper, we present a novel controller combining the variable admittance control and assistant control. In particular, the reinforcement learning is used to obtain the optimal damping value of the admittance controller by minimizing the reward function. In addition, we use the long short-term memory networks (LSTMs) to predict human intention based on the human limb dynamics and then an assistant controller is proposed to help human complete collaboration tasks. We validate the performance of our prediction algorithm and controller on a 7 d.o.f Franka Emika robot equipped with joint torque sensors. The proposed controller can both achieve minimum-jerk trajectory and low-effort cost.
{"title":"Human-Robot Collaboration using Variable Admittance Control and Human Intention Prediction","authors":"Wei-feng Lu, Zhe Hu, Jia Pan","doi":"10.1109/CASE48305.2020.9217040","DOIUrl":"https://doi.org/10.1109/CASE48305.2020.9217040","url":null,"abstract":"Due to the difficulty of modeling human limb, it is very challenging to design the controller for human-robot collaboration. In this paper, we present a novel controller combining the variable admittance control and assistant control. In particular, the reinforcement learning is used to obtain the optimal damping value of the admittance controller by minimizing the reward function. In addition, we use the long short-term memory networks (LSTMs) to predict human intention based on the human limb dynamics and then an assistant controller is proposed to help human complete collaboration tasks. We validate the performance of our prediction algorithm and controller on a 7 d.o.f Franka Emika robot equipped with joint torque sensors. The proposed controller can both achieve minimum-jerk trajectory and low-effort cost.","PeriodicalId":212181,"journal":{"name":"2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128389412","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：