Time filter

Source Type

Fang H.,Beijing Institute of Technology | Fang H.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Lu S.-L.,Beijing Institute of Technology | Lu S.-L.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | And 4 more authors.
Frontiers of Information Technology and Electronic Engineering | Year: 2017

Coalition formation is an important coordination problem in multi-agent systems, and a proper description of collaborative abilities for agents is the basic and key precondition in handling this problem. In this paper, a model of task-oriented collaborative abilities is established, where five task-oriented abilities are extracted to form a collaborative ability vector. A task demand vector is also described. In addition, a method of coalition formation with stochastic mechanism is proposed to reduce excessive competitions. An artificial intelligent algorithm is proposed to compensate for the difference between the expected and actual task requirements, which could improve the cognitive capabilities of agents for human commands. Simulations show the effectiveness of the proposed model and the distributed artificial intelligent algorithm. © 2017, Journal of Zhejiang University Science Editorial Office and Springer-Verlag Berlin Heidelberg.


Zhu H.,Beijing Institute of Technology | Zhu H.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Porikli F.,Australian National University
IEEE Transactions on Image Processing | Year: 2017

Tracking objects across multiple frames is a well-investigated problem in computer vision. The majority of the existing algorithms that assume an accurate initialization is readily available. However, in many real-life settings, in particular for applications where the video is streaming in real time, the initialization has to be provided by a human operator. This limitation raises an inevitable uncertainty issue. Here, we first collect a large and new data set of inputs that consists of more than 20 K human initialization clicks, by several subjects under three practical user interface scenarios for the popular TB50 tracking benchmark. We analyze the factors and mechanisms of human input, derive statistical models, and show that human input always contains deviations, which exacerbate further when the relative object-camera motion becomes large. We also design and evaluate alternative refinement schemes, and propose a strategy that refits an object window on the most probable target region after a single click. To compensate for the human initialization errors, our method generates window proposals using objectness cues extracted from color and motion attributes, accumulates them into a likelihood map that is weighted by the initial click position and visual saliency scores, and assigns the final window by the maximum likelihood estimate. Our experiments demonstrate that the presented refinement strategy effectively reduces human input errors. © 1992-2012 IEEE.


Chen J.,Beijing Institute of Technology | Chen J.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Zha W.,Beijing Institute of Technology | Zha W.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | And 3 more authors.
Automatica | Year: 2016

Inspired by the hunting and foraging behaviors of group predators, this paper addresses a class of multi-player pursuit-evasion games with one superior evader, who moves faster than the pursuers. We are concerned with the conditions under which the pursuers can capture the evader, involving the minimum number and initial spatial distribution required as well as the cooperative strategies of the pursuers. We present some necessary or sufficient conditions to regularize the encirclement formed by the pursuers to the evader. Then we provide a cooperative scheme for the pursuers to maintain and shrink the encirclement until the evader is captured. Finally, we give some examples to illustrate the theoretical results. © 2016 Elsevier Ltd. All rights reserved.


Qiao D.,Beijing Institute of Technology | Qiao D.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Guo Z.,Beijing Institute of Aerospace Control Devices | Ma H.,Beijing Institute of Technology | Shi S.,Sogou
Chinese Control Conference, CCC | Year: 2015

This paper proposes an iterative evolutionary algorithm with emulating nodes' local movement for searching the best localization accuracy in range-free scenario. All localization methods face a trade-off between the solution quality and computation cost. When sensor network localization in range-free scenario is considered as a constraint satisfaction problem, localization will reach the highest accuracy but with huge computation complexity. To solve the constraint satisfaction, some characteristics that only exist in range-free localization problem are utilized as heuristics in the search of nodes' positions. They are summarized as simple and complex movement to emulate nodes' local movement, and proved to be effective to find a suitable searching direction and jump out of local-minimums existing in the localization. Those emulations are then included in each iteration of a two-objective evolutionary algorithm minimizing the number of node-pairs with violated connectivity, as well as the value quantitating how worse of the violations. Simulation results show that the proposed algorithm can greatly decrease the reach high-accurate positions within limited iterations. © 2015 Technical Committee on Control Theory, Chinese Association of Automation.


Chao T.,Beijing Institute of Technology | Chao T.,State Key Laboratory of Intelligent Control and Decision of Complex Systems
Chinese Control Conference, CCC | Year: 2016

Alongside advancements in communication technology, Ad Hoc networks have grown considerably in popularity. In high-relative-velocity environments, however, air nodes often have supersonic velocity but ground nodes are nearly stationary. In this case, traditional protocols do not satisfy the necessary link stability though this type of node usually has clear purpose or mission. To this effect, node activity can be forecast to a certain extent. In this study, a motion model was developed to describe aircraft flying in high-relative-velocity environments. We also investigated methods of maximizing link lifetime and network stability by designing an improved LAR1 protocol. MatLab simulation results showed that the proposed LAR1 protocol has superior performances in terms of link lifetime, but may increase the number of hops in data transmission. © 2016 TCCT.


Hou Q.,Beijing Institute of Technology | Hou Q.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Chen W.,Beijing Institute of Technology | Chen W.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | And 2 more authors.
Chinese Control Conference, CCC | Year: 2016

In this paper, we propose an improved part-based tracking method based on compressive tracker. The traditional compressive tracking can hardly deal with occlusion and scale variation, which is not robust enough in these situations. In our part-based method, the occluded image patches which are selected by K-means are not able to participate in determining the location of the target in order to increase tracking accuracy. To solve the problem of scale variation, an algorithm is proposed to estimate the size of target with the information of the whole target. In addition, we also use the positions of every pail to help us to get a precise size of the target. With the combination of two methods, the scale of the target is obtained. Experiments compared with five state-of-the-arts have been done to prove the effectiveness and robustness of our method. © 2016 TCCT.


Yue Y.,Beijing Institute of Technology | Yue Y.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Song Z.,Beijing Institute of Technology | Song Z.,State Key Laboratory of Intelligent Control and Decision of Complex Systems
2015 IEEE International Conference on Information and Automation, ICIA 2015 - In conjunction with 2015 IEEE International Conference on Automation and Logistics | Year: 2015

Laser beam stabilization systems have important engineering applications, for example in aircraft/spacecraft targeting, and free space laser-based communication, etc. Recent studies show that such systems are negative-imaginary systems with free body motion. Based on negative-imaginary theory, this paper proposes an integral resonant controller for the Quanser's laser beam stabilization system and optimizes the controller parameters. The proposed control scheme not only compensates for vibration sources and stabilizes the laser beam position, but also guarantees the robustness with respect to system parameter uncertainty. Simulation results obtained from the Quanser's laser beam stabilization simulation model are reported and comparisons are made with a PID controller. © 2015 IEEE.


Dong Z.,Beijing Institute of Technology | Dong Z.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Zhang J.,Beijing Institute of Technology | Zhang J.,State Key Laboratory of Intelligent Control and Decision of Complex Systems
Chinese Control Conference, CCC | Year: 2015

Ball and plate system is a multivariable system with inherent nonlinearity. It is a two-dimensional extension of the classic ball and beam system. In this paper, the idea of Active Disturbance Rejection Control (ADRC) is introduced. The extended state observer (ESO) is applied to estimate the disturbance including nonlinearity and the friction. Then, the disturbance estimated can be eliminated from the control. For the sake of reducing the impact of the measurement noise, the Fal function filter is applied to filter the output of the system with measurement noise. Meanwhile, the feedforward compensator is designed to reduce the error of the system. In the end, the simulations of the set-point experiments and the circle trajectory tracking experiments are completed. The simulation results show that the ADRC controller combined with a feedforward compensator performances better. © 2015 Technical Committee on Control Theory, Chinese Association of Automation.


Duan C.,Beijing Institute of Technology | Duan C.,State Key Laboratory of Intelligent Control and Decision of Complex Systems
Chinese Control Conference, CCC | Year: 2015

Ample research has been conducted in routing protocols for Mobile Ad Hoc Networks while traditional protocols cannot satisfy the demand of highly-dynamic environment. Though nodes in highly-dynamic environment often have supersonic velocity, they mainly perform their task with the same mission and their velocity and direction are basically approximate. Hence, the feature makes it possible to forecast links' quality. In this paper, Gauss-Markov mobility model (GM) was introduced in order to reduce the influence of random movement on simulation performance; meanwhile we have improved the routing protocol based on DSR by utilizing the idea of link's lifetime. Although the improved protocol may increase the number of hops in data transmission, it makes the network more stable. Compared to the original DSR protocol based 'minimum hops', simulation results operated in Qualnet present us the superior performances of the improved protocol, especially in Packet Delivery Ratio, Received Throughput, Average End-to-end Delay and Average Jitter. © 2015 Technical Committee on Control Theory, Chinese Association of Automation.


Huang Y.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Wang J.,State Key Laboratory of Intelligent Control and Decision of Complex Systems | Shi D.,State Key Laboratory of Intelligent Control and Decision of Complex Systems
Chinese Control Conference, CCC | Year: 2016

In this work, an event-triggered high gain observer (ET-HGO) design is considered for a continuous-time nonlinear system combined with disturbance and uncertainty. In the event-trigger scheme, the performance of the ET-HGO depends on the triggering condition directly. Hence, in this work, for the high-gain observer considered, an event-triggered transmission strategy, which does not relies on the other system state but the output of the control system, is proposed such that the observation error is asymptotically bounded. Further, with mild restriction, the observation error is guaranteed to be bounded all the time. The obtained theoretical results are evaluated through the numerical simulations. © 2016 TCCT.

Loading State Key Laboratory of Intelligent Control and Decision of Complex Systems collaborators
Loading State Key Laboratory of Intelligent Control and Decision of Complex Systems collaborators