Pub Date : 2019-08-01DOI: 10.1109/COASE.2019.8843152
Mattias Hovgard, B. Lennartson, Kristofer Bengtsson
This paper presents energy optimization of a welding station from a manufacturing line in an automotive factory. The aim of the optimization is to find free time between operations, where it is possible to extend the execution time of the robot movements, and thereby saving energy, without extending the cycle time of the whole station. The station is modeled and optimized in a simulation platform. The optimization algorithm works by iteratively limiting the maximum velocity of the robot movements, until no more free time exists. Simulation results show that the energy use, peak power and jerk of the robots can be reduced significantly.
{"title":"Simulation Based Energy Optimization of Robot Stations by Motion Parameter Tuning","authors":"Mattias Hovgard, B. Lennartson, Kristofer Bengtsson","doi":"10.1109/COASE.2019.8843152","DOIUrl":"https://doi.org/10.1109/COASE.2019.8843152","url":null,"abstract":"This paper presents energy optimization of a welding station from a manufacturing line in an automotive factory. The aim of the optimization is to find free time between operations, where it is possible to extend the execution time of the robot movements, and thereby saving energy, without extending the cycle time of the whole station. The station is modeled and optimized in a simulation platform. The optimization algorithm works by iteratively limiting the maximum velocity of the robot movements, until no more free time exists. Simulation results show that the energy use, peak power and jerk of the robots can be reduced significantly.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"79 1","pages":"456-461"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90190012","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 : 2019-08-01DOI: 10.1109/COASE.2019.8843028
Chao-Bo Yan, Ziqian Zheng
High energy consumption is detrimental for reducing manufacturing cost and improving competitiveness of manufacturing enterprises. In this paper, the problem of reducing energy consumption in a three-machine Bernoulli serial line is formulated and solved. Specifically, first, based on the aggregation method, structural characteristics of the optimization model are analyzed and thus, the problem is transformed; then, with the efficiency of the first machine fixed, based on the results of the energy consumption optimization model in the two-machine line, the optimization model is further analyzed and solved; finally, the property of the objective function with respect to the efficiency of the first machine is analyzed and based on the property, an algorithm is designed to solve the energy consumption optimization problem in the three-machine Bernoulli serial line.
{"title":"Energy Consumption Optimization in Three-Machine Bernoulli Serial Lines","authors":"Chao-Bo Yan, Ziqian Zheng","doi":"10.1109/COASE.2019.8843028","DOIUrl":"https://doi.org/10.1109/COASE.2019.8843028","url":null,"abstract":"High energy consumption is detrimental for reducing manufacturing cost and improving competitiveness of manufacturing enterprises. In this paper, the problem of reducing energy consumption in a three-machine Bernoulli serial line is formulated and solved. Specifically, first, based on the aggregation method, structural characteristics of the optimization model are analyzed and thus, the problem is transformed; then, with the efficiency of the first machine fixed, based on the results of the energy consumption optimization model in the two-machine line, the optimization model is further analyzed and solved; finally, the property of the objective function with respect to the efficiency of the first machine is analyzed and based on the property, an algorithm is designed to solve the energy consumption optimization problem in the three-machine Bernoulli serial line.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"31 1","pages":"590-595"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90489856","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 : 2019-08-01DOI: 10.1109/COASE.2019.8843223
Sayyed Jaffar Ali Raza, Mingjie Lin
Policy based reinforcement learning methods are widely used for multi-agent systems to learn optimal actions given any state; with partial or even no model representation. However multi-agent systems with complex structures (curse of dimensionality) or with high constraints (like bio-inspired (a) snake or serpentine robots) show limited performance in such environments due to sparse-reward nature of environment and no fully observable model representation. In this paper we present a constructive learning and planning scheme that reduces the complexity of high-diemensional agent model by decomposing it into identical, connected and scaled down multiagent structure and then apply learning framework in layers of local and global ranking. Our layered hierarchy method also decomposes the final goal into multiple sub-tasks and a global task (final goal) that is bias-induced function of local sub-tasks. Local layer deals with learning ‘reusable’ local policy for a local agent to achieve a sub-task optimally; that local policy can also be reused by other identical local agents. Furthermore, global layer learns a policy to apply right combination of local policies that are parameterized over entire connected structure of local agents to achieve the global task by collaborative construction of local agents. After learning local policies and while learning global policy, the framework generates sub-tasks for each local agent, and accepts local agents’ intrinsic rewards as positive bias towards maximum global reward based of optimal sub-tasks assignments. The advantage of proposed approach includes better exploration due to decomposition of dimensions, and reusability of learning paradigm over extended dimension spaces. We apply the constructive policy method to serpentine robot with hyper-redundant degrees of freedom (DOF), for achieving optimal control and we also outline connection to hierarchical apprenticeship learning methods which can be seen as layered learning framework for complex control tasks.
{"title":"Constructive Policy: Reinforcement Learning Approach for Connected Multi-Agent Systems","authors":"Sayyed Jaffar Ali Raza, Mingjie Lin","doi":"10.1109/COASE.2019.8843223","DOIUrl":"https://doi.org/10.1109/COASE.2019.8843223","url":null,"abstract":"Policy based reinforcement learning methods are widely used for multi-agent systems to learn optimal actions given any state; with partial or even no model representation. However multi-agent systems with complex structures (curse of dimensionality) or with high constraints (like bio-inspired (a) snake or serpentine robots) show limited performance in such environments due to sparse-reward nature of environment and no fully observable model representation. In this paper we present a constructive learning and planning scheme that reduces the complexity of high-diemensional agent model by decomposing it into identical, connected and scaled down multiagent structure and then apply learning framework in layers of local and global ranking. Our layered hierarchy method also decomposes the final goal into multiple sub-tasks and a global task (final goal) that is bias-induced function of local sub-tasks. Local layer deals with learning ‘reusable’ local policy for a local agent to achieve a sub-task optimally; that local policy can also be reused by other identical local agents. Furthermore, global layer learns a policy to apply right combination of local policies that are parameterized over entire connected structure of local agents to achieve the global task by collaborative construction of local agents. After learning local policies and while learning global policy, the framework generates sub-tasks for each local agent, and accepts local agents’ intrinsic rewards as positive bias towards maximum global reward based of optimal sub-tasks assignments. The advantage of proposed approach includes better exploration due to decomposition of dimensions, and reusability of learning paradigm over extended dimension spaces. We apply the constructive policy method to serpentine robot with hyper-redundant degrees of freedom (DOF), for achieving optimal control and we also outline connection to hierarchical apprenticeship learning methods which can be seen as layered learning framework for complex control tasks.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"23 1","pages":"257-262"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83211457","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 : 2019-08-01DOI: 10.1109/COASE.2019.8843323
S. Roselli, Fredrik Hagebring, Sarmad Riazi, Martin Fabian, K. Åkesson
Bin covering is an important optimization problem in many industrial fields, such as packaging, recycling, and food processing. The problem concerns a set of items, each with its own value, that are to be collected into bins in such a way that the total value of each bin, as measured by the sum of its item values, is not lower than a target value. The optimization problem concerns maximizing the number of bins. This is a combinatorial NP-hard problem, for which true optimal solutions can only be calculated in specific cases, such as when restricted to a small number of items. To get around this problem, many suboptimal approaches exist. This paper describes a formulation of the bin covering that allows to find the true optimum for a rather large number of items, over 1000. Also presented is a suboptimal solution, which is compared to the true optimum and found to come within less than 10% of the optimum.
{"title":"On the Use of Equivalence Classes for Optimal and Sub-Optimal Bin Covering","authors":"S. Roselli, Fredrik Hagebring, Sarmad Riazi, Martin Fabian, K. Åkesson","doi":"10.1109/COASE.2019.8843323","DOIUrl":"https://doi.org/10.1109/COASE.2019.8843323","url":null,"abstract":"Bin covering is an important optimization problem in many industrial fields, such as packaging, recycling, and food processing. The problem concerns a set of items, each with its own value, that are to be collected into bins in such a way that the total value of each bin, as measured by the sum of its item values, is not lower than a target value. The optimization problem concerns maximizing the number of bins. This is a combinatorial NP-hard problem, for which true optimal solutions can only be calculated in specific cases, such as when restricted to a small number of items. To get around this problem, many suboptimal approaches exist. This paper describes a formulation of the bin covering that allows to find the true optimum for a rather large number of items, over 1000. Also presented is a suboptimal solution, which is compared to the true optimum and found to come within less than 10% of the optimum.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"29 8 1","pages":"1004-1009"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83531831","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 : 2019-08-01DOI: 10.1109/COASE.2019.8843136
Lingbo Cheng, Mahdi Tavakoli
A neural-network-based heart motion prediction method is proposed for ultrasound-guided beating-heart surgery to compensate for time delays caused by ultrasound (US) image acquisition and processing. Such image processing is needed for tracking heart tissue in US images, which is itself a requirement for beating-heart surgery. Once the heart tissue is tracked in US images, a recurrent neural network (NN) is employed to learn how to predict the motion of the tracked heart motion in order to compensate for the delays introduced in the initial US image processing step. To verify the feasibility of predicting both simple and complex heart motions, the NN is tested with two types of heart motion data: (i) fixed heart rate and maximum amplitude, and (ii) varying heart rate and maximum amplitude. Also, the NN was tested for different prediction horizons and showed effectiveness for both small and large delays. The heart motion prediction results using NN are compared to the results using an extended Kalman filter (EKF) algorithm. Using NN, the mean absolute error and the root mean squared error between the predicted and the actually tracked heart motions are roughly 60% smaller than those achieved by using the EKF. Moreover, the NN is able to predict the heart position up to 1000 ms in advance, which significantly exceeds the typical US image acquisition/processing delays for this application (160 ms in our tests). Overall, the NN predictor shows significant advantages (higher accuracy and longer prediction horizon) compared to the EKF predictor.
{"title":"Neural-Network-Based Heart Motion Prediction for Ultrasound-Guided Beating-Heart Surgery","authors":"Lingbo Cheng, Mahdi Tavakoli","doi":"10.1109/COASE.2019.8843136","DOIUrl":"https://doi.org/10.1109/COASE.2019.8843136","url":null,"abstract":"A neural-network-based heart motion prediction method is proposed for ultrasound-guided beating-heart surgery to compensate for time delays caused by ultrasound (US) image acquisition and processing. Such image processing is needed for tracking heart tissue in US images, which is itself a requirement for beating-heart surgery. Once the heart tissue is tracked in US images, a recurrent neural network (NN) is employed to learn how to predict the motion of the tracked heart motion in order to compensate for the delays introduced in the initial US image processing step. To verify the feasibility of predicting both simple and complex heart motions, the NN is tested with two types of heart motion data: (i) fixed heart rate and maximum amplitude, and (ii) varying heart rate and maximum amplitude. Also, the NN was tested for different prediction horizons and showed effectiveness for both small and large delays. The heart motion prediction results using NN are compared to the results using an extended Kalman filter (EKF) algorithm. Using NN, the mean absolute error and the root mean squared error between the predicted and the actually tracked heart motions are roughly 60% smaller than those achieved by using the EKF. Moreover, the NN is able to predict the heart position up to 1000 ms in advance, which significantly exceeds the typical US image acquisition/processing delays for this application (160 ms in our tests). Overall, the NN predictor shows significant advantages (higher accuracy and longer prediction horizon) compared to the EKF predictor.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"52 4 1","pages":"437-442"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83541215","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 : 2019-08-01DOI: 10.1109/COASE.2019.8842849
Sarmad Riazi, Thomas Diding, P. Falkman, Kristofer Bengtsson, B. Lennartson
In this paper, we propose a new heuristic as well as several improvements to an existing approach based on Benders decomposition for solving the conflict free scheduling and routing of automated guided vehicles (AGVs), with promising results. The existing method solves the problem in two stages; task assignment/sequencing, and feasibility check of the first stage’s solution subject to collision-avoidance constraints. The method is not suitable for large-scale AGV systems. We proposed several improvements and speedup strategies that result in fast methods capable of scheduling AGVs in a realistic layout with a graph of several hundred nodes and arcs. This is done by reformulating the mathematical model of the problem. We also introduce a new heuristic based on the improved method that yields high-quality solutions quickly. Moreover, we solve a real large-scale industrial instance by a commercial constraint programming solver and an open-source SMT solver. The results show that both of these general-purpose solvers can effectively solve our proposed models.
{"title":"Scheduling and Routing of AGVs for Large-scale Flexible Manufacturing Systems","authors":"Sarmad Riazi, Thomas Diding, P. Falkman, Kristofer Bengtsson, B. Lennartson","doi":"10.1109/COASE.2019.8842849","DOIUrl":"https://doi.org/10.1109/COASE.2019.8842849","url":null,"abstract":"In this paper, we propose a new heuristic as well as several improvements to an existing approach based on Benders decomposition for solving the conflict free scheduling and routing of automated guided vehicles (AGVs), with promising results. The existing method solves the problem in two stages; task assignment/sequencing, and feasibility check of the first stage’s solution subject to collision-avoidance constraints. The method is not suitable for large-scale AGV systems. We proposed several improvements and speedup strategies that result in fast methods capable of scheduling AGVs in a realistic layout with a graph of several hundred nodes and arcs. This is done by reformulating the mathematical model of the problem. We also introduce a new heuristic based on the improved method that yields high-quality solutions quickly. Moreover, we solve a real large-scale industrial instance by a commercial constraint programming solver and an open-source SMT solver. The results show that both of these general-purpose solvers can effectively solve our proposed models.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"111 1","pages":"891-896"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84781861","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 : 2019-08-01DOI: 10.1109/COASE.2019.8842895
Biao Hu, Zhengcai Cao, Lijie Zhou
In this paper, we first propose laxity-based strategy that prioritizes applications based on the laxity of meeting their deadlines. Application with the least laxity will have the highest scheduling priority. Calculating the application laxity will consume some computation time, which may not be practicable for the online implementation. To overcome this problem, we further propose transferring the application deadline to its inside tasks, which makes the laxity calculation easier. We also apply the laxity-based scheduling algorithm to schedule applications with multiple criticalities. Towards the challenge of reconciling timing requirements from different criticality applications, system mode-switch scheme and virtual deadlines are adopted to preferentially guarantee high-critical applications when system is overloaded. Experimental results demonstrate that on the one hand our proposed algorithms can greatly reduce the deadline misses, and on the other hand timing requirements of high-critical applications can be more stringently guaranteed compared with low-critical applications.
{"title":"Adaptive Real-Time Scheduling of Dynamic Multiple-Criticality Applications on Heterogeneous Distributed Computing Systems","authors":"Biao Hu, Zhengcai Cao, Lijie Zhou","doi":"10.1109/COASE.2019.8842895","DOIUrl":"https://doi.org/10.1109/COASE.2019.8842895","url":null,"abstract":"In this paper, we first propose laxity-based strategy that prioritizes applications based on the laxity of meeting their deadlines. Application with the least laxity will have the highest scheduling priority. Calculating the application laxity will consume some computation time, which may not be practicable for the online implementation. To overcome this problem, we further propose transferring the application deadline to its inside tasks, which makes the laxity calculation easier. We also apply the laxity-based scheduling algorithm to schedule applications with multiple criticalities. Towards the challenge of reconciling timing requirements from different criticality applications, system mode-switch scheme and virtual deadlines are adopted to preferentially guarantee high-critical applications when system is overloaded. Experimental results demonstrate that on the one hand our proposed algorithms can greatly reduce the deadline misses, and on the other hand timing requirements of high-critical applications can be more stringently guaranteed compared with low-critical applications.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"16 1","pages":"897-903"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87396120","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 : 2019-08-01DOI: 10.1109/COASE.2019.8843014
Xiaojie Wang, Yongpei Guan, Xiang Zhong
A new type of clinical visits-video-conferencing (VC) meeting with nurse assistance is emerging. Such virtual care services provide convenience and reduce travel burden of patients; however, it is unclear whether both patients and the healthcare organization would benefit from this service, due to the cost incurred by coordinating nurses to assist patients at home. Motivated by the Veterans Affairs (VA) Telehealth program, in this paper, we consider a healthcare organization offering both in-person and VC visits to community patients within its catchment area. Communities are characterized by distance and population density. A revenue-maximizing problem is formulated to identify the best pricing and patient diversion strategy that is also incentive compatible for patients. Our results show that under certain cost structures, both patients and the healthcare organization will be strictly better off with VC visits. The insights obtained from this work would support the design and implementation of VC visits that maximize the potential of telehealth to improve patient access.
{"title":"Design of Video-Based Clinical Visits with Nurse Assistant for Chronic Diseases Management","authors":"Xiaojie Wang, Yongpei Guan, Xiang Zhong","doi":"10.1109/COASE.2019.8843014","DOIUrl":"https://doi.org/10.1109/COASE.2019.8843014","url":null,"abstract":"A new type of clinical visits-video-conferencing (VC) meeting with nurse assistance is emerging. Such virtual care services provide convenience and reduce travel burden of patients; however, it is unclear whether both patients and the healthcare organization would benefit from this service, due to the cost incurred by coordinating nurses to assist patients at home. Motivated by the Veterans Affairs (VA) Telehealth program, in this paper, we consider a healthcare organization offering both in-person and VC visits to community patients within its catchment area. Communities are characterized by distance and population density. A revenue-maximizing problem is formulated to identify the best pricing and patient diversion strategy that is also incentive compatible for patients. Our results show that under certain cost structures, both patients and the healthcare organization will be strictly better off with VC visits. The insights obtained from this work would support the design and implementation of VC visits that maximize the potential of telehealth to improve patient access.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"42 1","pages":"709-714"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80666063","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 : 2019-08-01DOI: 10.1109/COASE.2019.8843050
Muhammad Usman Khalid, Janik M. Hager, W. Kraus, Marco F. Huber, Marc Toussaint
For most industrial bin picking solutions, the pose of a workpiece is localized by matching a CAD model to point cloud obtained from 3D sensor. Distinguishing flat workpieces from bottom of the bin in point cloud imposes challenges in the localization of workpieces that lead to wrong or phantom detections. In this paper, we propose a framework that solves this problem by automatically segmenting workpiece regions from non-workpiece regions in a point cloud data. It is done in real time by applying a fully convolutional neural network trained on both simulated and real data. The real data has been labelled by our novel technique which automatically generates ground truth labels for real point clouds. Along with real time workpiece segmentation, our framework also helps in improving the number of detected workpieces and estimating the correct object poses. Moreover, it decreases the computation time by approximately 1s due to a reduction of the search space for the object pose estimation.
{"title":"Deep Workpiece Region Segmentation for Bin Picking","authors":"Muhammad Usman Khalid, Janik M. Hager, W. Kraus, Marco F. Huber, Marc Toussaint","doi":"10.1109/COASE.2019.8843050","DOIUrl":"https://doi.org/10.1109/COASE.2019.8843050","url":null,"abstract":"For most industrial bin picking solutions, the pose of a workpiece is localized by matching a CAD model to point cloud obtained from 3D sensor. Distinguishing flat workpieces from bottom of the bin in point cloud imposes challenges in the localization of workpieces that lead to wrong or phantom detections. In this paper, we propose a framework that solves this problem by automatically segmenting workpiece regions from non-workpiece regions in a point cloud data. It is done in real time by applying a fully convolutional neural network trained on both simulated and real data. The real data has been labelled by our novel technique which automatically generates ground truth labels for real point clouds. Along with real time workpiece segmentation, our framework also helps in improving the number of detected workpieces and estimating the correct object poses. Moreover, it decreases the computation time by approximately 1s due to a reduction of the search space for the object pose estimation.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"35 1","pages":"1138-1144"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88171987","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 : 2019-08-01DOI: 10.1109/COASE.2019.8842927
Haitao Yuan, J. Bi, Mengchu Zhou
Distributed clouds (DCs) often require a huge amount of energy to provide multiple services to users around the world. Users bring revenue to DC providers based on the quality of service (QoS) of tasks. These tasks are transmitted to DCs through many available Internet service providers (ISPs) with different bandwidth prices and capacities. Besides, power grid prices, and green energy in different DCs differ with different geographical sites. Consequently, it is challenging to execute tasks among DCs in a high-QoS and high-profit way. This work proposes a bi-objective optimization algorithm to maximize the profit of a DC provider, and minimize the loss possibility of all tasks by specifying the allocation of tasks among different ISPs, and task service rates of each DC. A constrained optimization problem is given and solved by a novel Simulated-annealing-based Bi-objective Differential Evolution (SBDE) algorithm to produce a close-to-optimal Pareto set of solutions. The minimum Manhattan distance is further used to obtain a knee solution, and it determines Pareto optimal service rates and task allocation among ISPs. Realistic trace-driven results demonstrate that SBDE realizes less loss possibility of tasks, and higher profit than several state-of-the-art scheduling algorithms.
{"title":"QoS and Profit Aware Task Scheduling with Simulated-Annealing-Based Bi-Objective Differential Evolution in Green Clouds","authors":"Haitao Yuan, J. Bi, Mengchu Zhou","doi":"10.1109/COASE.2019.8842927","DOIUrl":"https://doi.org/10.1109/COASE.2019.8842927","url":null,"abstract":"Distributed clouds (DCs) often require a huge amount of energy to provide multiple services to users around the world. Users bring revenue to DC providers based on the quality of service (QoS) of tasks. These tasks are transmitted to DCs through many available Internet service providers (ISPs) with different bandwidth prices and capacities. Besides, power grid prices, and green energy in different DCs differ with different geographical sites. Consequently, it is challenging to execute tasks among DCs in a high-QoS and high-profit way. This work proposes a bi-objective optimization algorithm to maximize the profit of a DC provider, and minimize the loss possibility of all tasks by specifying the allocation of tasks among different ISPs, and task service rates of each DC. A constrained optimization problem is given and solved by a novel Simulated-annealing-based Bi-objective Differential Evolution (SBDE) algorithm to produce a close-to-optimal Pareto set of solutions. The minimum Manhattan distance is further used to obtain a knee solution, and it determines Pareto optimal service rates and task allocation among ISPs. Realistic trace-driven results demonstrate that SBDE realizes less loss possibility of tasks, and higher profit than several state-of-the-art scheduling algorithms.","PeriodicalId":6695,"journal":{"name":"2019 IEEE 15th International Conference on Automation Science and Engineering (CASE)","volume":"21 1","pages":"904-909"},"PeriodicalIF":0.0,"publicationDate":"2019-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91170982","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：