Papers by SATOSHI HOSHINO
Journal of the Robotics Society of Japan
We have thus far presented a brain-machine-interface, BMI, for users of personal mobility robots.... more We have thus far presented a brain-machine-interface, BMI, for users of personal mobility robots. However, once the BMI predicts a wrong control command, both the user and robot face the danger of collision accidents. In this paper, therefore, we propose a fail-safe controller based on CNN (Convolutional Neural Network) for assisting users of personal mobility robots with the BMI. In addition to the control command, a depth map for the input image is simultaneously predicted by the fail-safe controller through multi-task learning. For this purpose, CAE (Convolutional Autoencoder) and DCGAN (Deep Convolutional Generative Adversarial Networks) are used instead of the CNN. In the experiments, we show that the fail-safe performance is increased by predicting the depth map for the input image. Finally, the fail-safe controller based on the DCGAN yields the best performance.
Journal of the Robotics Society of Japan, 2013
In order for mobile robots to move in convoy, a small platooning technique based on adaptive crui... more In order for mobile robots to move in convoy, a small platooning technique based on adaptive cruise control (ACC) and a simple clustering method among adjacent robots have been proposed by the author. In the previous works, however, homogeneous mobile robots have been assumed in terms of controllers. In this paper, on the other hand, heterogeneous mobile robots depending on different controllers are assumed. For these robots, a more powerful small platooning technique focusing on a velocity distribution is proposed. In the small platooning methodology, variable time headway in ACC model and dynamic environmental rule are presented. The former is for adjusting a time headway of robots in a congested segment in a lane according to the congestion degree, and the latter is provided into a bottleneck lane where the robots move most slowly so as to externally regulate the behavior. In simulation experiments, the previous and proposed small platooning techniques are applied to heterogeneous mobile robots, and finally, the effectiveness of the proposed small platooning technique is shown.
Journal of the Robotics Society of Japan, 2013
In order for mobile robots to move in convoy, a small platooning technique based on adaptive crui... more In order for mobile robots to move in convoy, a small platooning technique based on adaptive cruise control (ACC) and a simple clustering method among adjacent robots have been proposed by the author. In the previous works, however, homogeneous mobile robots have been assumed in terms of controllers. In this paper, on the other hand, heterogeneous mobile robots depending on different controllers are assumed. For these robots, a more powerful small platooning technique focusing on a velocity distribution is proposed. In the small platooning methodology, variable time headway in ACC model and dynamic environmental rule are presented. The former is for adjusting a time headway of robots in a congested segment in a lane according to the congestion degree, and the latter is provided into a bottleneck lane where the robots move most slowly so as to externally regulate the behavior. In simulation experiments, the previous and proposed small platooning techniques are applied to heterogeneous mobile robots, and finally, the effectiveness of the proposed small platooning technique is shown.
IFAC Proceedings Volumes, 2013
A system state is changed to congestion as the number of autonomous mobile robots is increased. E... more A system state is changed to congestion as the number of autonomous mobile robots is increased. Eventually, jams of robots are prone to be formed. Jams have a profound influence on robots' behavior. Even in this state, therefore, robots are required to solve jams and move efficiently. For this challenge, adaptive cruise control (ACC) suitable for a vehicle platoon is applied to robots. For the platooning robots based on ACC, a reactive clustering method using local information among adjacent robots is proposed. These robots are thus clustered into small platoons responding to a variety of changing circumstances. Through simulation experiments, several scenarios in order for robots to organize clusters are compared and the reasonability of the clustering method for the robot platooning is evaluated. Furthermore, the effectiveness of the clustering method is discussed in comparison to a robot coordination methodology that requires global information.
2005 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2005
The development of a highly efficient management methodology for an Automated Container Terminal ... more The development of a highly efficient management methodology for an Automated Container Terminal (ACT) poses a problem for port authorities. The focus here is on a transportation system for an Automated Guided Vehicle (AGV) for an ACT. In this paper, we design the detailed management models, i.e., agent cooperation and container storage planning for the transportation system. Then, we optimally design systems that are constructed with the use of the designed management models. Comparisons of the systems are made to evaluate cost effectiveness based on the total construction cost and validity of the management models. Finally, a proposal is made for the most efficient management system.
2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566)
In this paper, we provide the comparison indicator of the AGV transportation systems. For this pu... more In this paper, we provide the comparison indicator of the AGV transportation systems. For this purpose, we propose an optimal design methodology for the AGV transportation system by using the queuing network theory. In this methodology, the queuing network theory and a simulation-based optimization method are integrated to obtain the optimal design parameters (i.e., these are the design solutions of this design problem). In this study, two different types of the AGV transportation systems are designed. Then the performance of transportation is compared. Finally, the characteristics of each transportation system that depends on the design parameters is provided.
Robot Vision System for Real-Time Human Detection and Action Recognition
Intelligent Autonomous Systems 15, 2018
Mobile robots equipped with camera sensors are required to perceive surrounding humans and their ... more Mobile robots equipped with camera sensors are required to perceive surrounding humans and their actions for safe autonomous navigation. These are so-called human detection and action recognition. In this paper, moving humans are target objects. Compared to computer vision, the real-time performance of robot vision is more important. For this challenge, we propose a robot vision system. In this system, images described by the optical flow are used as an input. For the classification of humans and actions in the input images, we use Convolutional Neural Network, CNN, rather than coding invariant features. Moreover, we present a novel detector, local search window, for clipping partial images around target objects. Through the experiment, finally, we show that the robot vision system is able to detect the moving human and recognize the action in real time.
Journal of the Robotics Society of Japan, 2012
Safe and efficient motion planning of multiple mobile robots based on artificial potential for human behavior and robot congestion
Advanced Robotics, 2015
ABSTRACT In order for robots to exist together with humans, safety for the humans has to be stric... more ABSTRACT In order for robots to exist together with humans, safety for the humans has to be strictly ensured. On the other hand, safety might decrease working efficiency of robots. Namely, this is a trade-off problem between human safety and robot efficiency in a field of human-robot interaction. For this problem, we propose a novel motion planning technique of multiple mobile robots. Two artificial potentials are presented for generating repulsive force. The first potential is provided for humans. The von Mises distribution is used to consider the behavioral property of humans. The second potential is provided for the robots. The Kernel density estimation is used to consider the global robot congestion. Through simulation experiments, the effectiveness of the behavior and congestion potentials of the motion planning technique for human safety and robot efficiency is discussed. Moreover, a sensing system for humans in a real environment is developed. From experimental results, the significance of the behavior potential based on the actual humans is discussed. For the coexistence of humans and robots, it is important to evaluate a mutual influence between them. For this purpose, a virtual space is built using projection mapping. Finally, the effectiveness of the motion planning technique for the human-robot interaction is discussed from the point of view of not only robots but also humans.
2007 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2007
For the design of an automated transportation system in an actual seaport container terminal, it ... more For the design of an automated transportation system in an actual seaport container terminal, it is necessary to take into consideration the maintenance of operating robots (AGV: Automated Guided Vehicle and ATC: Automated Transfer Crane). For this purpose, we develop an operation model in which each robot enters a maintenance mode while operating on the basis of its reliability. We also aim to design the mean time between failure (MTBF) which is a kind of a measure of robot performance on the basis of the robots' reliability as well as the number of robots. However, this is a combinatorial design problem. Therefore, we propose a design methodology in order to derive one combinatorial design solution for a given demand by considering a system management cost that includes a system construction cost and a penalty cost. The designed systems are evaluated on the basis of the system management costs. Finally, we present the validity of the proposed design methodology and designed systems.
IFAC Proceedings Volumes, 2008
In this paper, we address a fluctuating low-volume and high-mix manufacturing system that handles... more In this paper, we address a fluctuating low-volume and high-mix manufacturing system that handles a large variety of products. The incoming products have their own task information imposed by customer demands. In other words, they are not given in advance. Consequently, the system-operating conditions vary with the changes in the given tasks over time. Because of such unpredictable fluctuations, the system might have localized heavy workloads. Therefore, workload balancing is a challenge. For this issue, we focus on multi-robots in a manufacturing system. We first describe behavioral and cooperation mechanisms among robots. Thus, by sharing their information, robots are able to respond to such dynamically changing situations reactively. We then develop a flexible and agile multi-robot manufacturing system with Automated Guided Vehicles (AGVs) and product-processing robots. Finally, we discuss the validity of the developed system.
IFAC Proceedings Volumes, 2008
Recycle systems exhibit steady state input multiplicities due to interactions of the units, even ... more Recycle systems exhibit steady state input multiplicities due to interactions of the units, even though individual units are quite simple. When handling constraints, if prospective constraint variables show such nonlinearity, control problems may arise because the steady state gain changes its sign. Using the reactor/separator system with two material recycles as a process example, a robust constraint handling controller is designed by confining the input into the large gain directions. Such directions are obtained as the Pareto optimal front of the multi-objective optimization problem which minimizes energy consumption in each unit. Performance of the designed controller is demonstrated through simulations.
2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2011
This paper focuses on the effectiveness of maintenance in fault-tolerant multi-robot systems. Suc... more This paper focuses on the effectiveness of maintenance in fault-tolerant multi-robot systems. Such a system is enabled to work as long as one robot works; namely, completely parallel. For this system, it is required to ensure fault tolerance and maintain high performance taking robot failures and maintenance, such as prevention and correction of robots, into account. Therefore, we propose an optimal maintenance strategy on the basis of reliability engineering. This strategy enables robots to undergo preventive maintenance at optimal intervals and corrective maintenance each time they fail. Through simulation experiments, the effectiveness of the optimal maintenance strategy is investigated. In addition, the influence of robots undergoing maintenance on system performance is mathematically and experimentally discussed on the basis of the number of robots and maintenance strategies.
Multi-robot coordination methodology in congested systems with bottlenecks
2011 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2011
ABSTRACT This paper describes a novel coordination methodology for autonomous mobile robots in co... more ABSTRACT This paper describes a novel coordination methodology for autonomous mobile robots in congested systems with bottlenecks. This methodology consists of two approaches. 1) A previous robot behavior control technique that utilizes external interaction force among robots is improved. This enables robots to reduce velocity not only for the jam preceding them, but also for the decelerating robot immediately in front. 2) An environmental rule in connection with the interaction force is designed and provided on congested lanes where robots move slowly. Thus, amplified interaction force affects the robots, and they move more slowly in the congested lanes. The improved interaction force and environmental rule are implemented in simulation experiments and compared to the previous robot behavior control technique and an adaptive cruise control (ACC) that has been proposed for autonomous vehicles. The interaction force and environmental rule can be implemented in the ACC; then, the resulting ACC improvement is discussed. Finally, the effectiveness of the usage of the improved interaction force and environmental rule for multi-robot coordination in congested systems with bottlenecks is shown.
2008 IEEE International Conference on Automation Science and Engineering, 2008
An off-site system in a petroleum refining plant mainly consists of storage tanks, gasoline compo... more An off-site system in a petroleum refining plant mainly consists of storage tanks, gasoline component blenders, and a pipe network that connects tanks with blenders. Therefore, in order to produce a required gasoline on time, these equipments have to be selected and used appropriately at the right time. However, previous studies have not taken into account the pipe network. In this paper, a novel scheduling model with the MILP method for gasoline blending in consideration of storage tanks and a pipe network is proposed. In order to model and formulate the pipe network and constraints, a previouslyproposed scheduling model is extended with the use of the set theory. In this regard, two modeling techniques are introduced to reduce a computational cost. Through a case study of the gasoline-blending scheduling, the effectiveness of the proposed and integrated scheduling model is finally shown.
Journal of Mechanical Systems for Transportation and Logistics, 2012
In this paper, cooperative behavior of autonomous mobile robots against congestion is experimenta... more In this paper, cooperative behavior of autonomous mobile robots against congestion is experimentally analyzed through multi-robot simulations. For this purpose, robots behavior regarding velocity is investigated. Two circular systems in which bottlenecks do not exist and exist are the research objects. In the simulation experiments, first off, a simple behavior model is applied to the robots. From the analysis results, behavioral features of the robots depending on two states, i.e., non-congestion and congestion, and presence or absence of bottlenecks are presented. These are criteria for evaluating the effectiveness of behavior models. After that, a proposed behavior model is applied to the robots and the effectiveness of cooperative behavior of the robots is evaluated on the basis of the criteria. Finally, it is shown that the proposed behavior model effectively solves jams or alleviates the influence in the congested systems regardless of bottlenecks.
Journal of Mechanical Systems for Transportation and Logistics, 2010
A container terminal consists of many agents. They include gantry cranes, transfer cranes (TCs), ... more A container terminal consists of many agents. They include gantry cranes, transfer cranes (TCs), internal trucks, and external trucks. Many containers are moved from the coastal terminals of Japan to inland sites and vice versa. The workload distribution among storage blocks (container stacking areas) changes over time in a container terminal. Therefore, TCs, which are mobile cranes for container handling in storage blocks, have to move appropriately among blocks according to the distribution, and the efficiency of yard operations depends heavily on the productivity of TCs. In the proposed method, we use a look-ahead methodology to determine the dynamic deployment of TCs, in which the workload distribution is estimated with simulated operations of TCs and adequate movement of TCs is obtained with searching methodology in real time. The simulation results from 10 hours of operation with 600 containers show that the proposed method can improve the efficiency of yard operations (average waiting time of trucks) by 65% in comparison to cases without movement of TCs and by 42% in comparison to the simple rule-based method. These results demonstrate the effectiveness of the proposed method.
2011 IEEE International Conference on Robotics and Automation, 2011
Ensuring fault tolerance of robotic systems is a challenge for factory automation. This paper foc... more Ensuring fault tolerance of robotic systems is a challenge for factory automation. This paper focuses on multirobot manipulation and maintenance for fault-tolerant systems. For this purpose, a manipulation strategy of multiple mobile robots, that enables a system to continue operating even if a working robot undergoes preventive maintenance or fails and undergoes corrective maintenance, is implemented. In addition, a robot failure and a maintenance policy for preventive and corrective maintenance activities are mathematically modeled on the basis of reliability engineering. Thus, working robots are allowed to undergo preventive maintenance at an optimal interval and corrective maintenance each time they fail. Finally, through simulation experiments, the effectiveness of an integrated multi-robot manipulation and maintenance in industrial applications is shown.
Proceedings 2007 IEEE International Conference on Robotics and Automation, 2007
Automation of transportation systems and promotion of the operations are becoming an internationa... more Automation of transportation systems and promotion of the operations are becoming an international demand on seaport container terminals. For this issue, we propose an integrated design methodology. In other words, in addition to the design of the appropriate number of machines, efficient system layout, and system management models, we attempt to design machines' specifications, which are operating in an automated transportation system. The objective of this study is to maximize the system efficiency while minimizing the changes of the specifications. Therefore, we evaluate the specifications based on the system throughput of the constructed system; then, make clear the impacts of the specifications on the system. Finally, for imposed demands, we design highly efficient transportation systems. From the construction costs and design results, we show a need to take into consideration the machines' specifications.
2007 IEEE International Conference on Automation Science and Engineering, 2007
In order to provide a highly efficient containerhandling system in a seaport terminal, for a give... more In order to provide a highly efficient containerhandling system in a seaport terminal, for a given constraint, i.e., demand, we take into account the performance of operating robots, such as quay container cranes (QCCs), automated guided vehicles (AGVs), and rubber-tired gantry cranes (RT-GCs) in addition to the number of robots, as design objectives. However, this is a combinatorial design problem. Therefore in this paper, we propose a design methodology with the use of a hybrid design process which had been proposed for designing the number of robots. By using this methodology, we design two objectives, that is, the number of robots and their performance for a given demand. Finally, we present the validity of the proposed methodology by comparing and evaluating construction costs of two systems, which are designed with the used of the proposed design methodology and a design methodology which does not take into account the robots performance. Moreover, we discuss the designed robots performance in terms of the system bottleneck.
Uploads
Papers by SATOSHI HOSHINO