With the development of deep neural networks, the model of restricted Boltzmann machine(RBM) has gradually become one of the essential aspects in deep learning researches. Because of the presence of the partition func...
详细信息
We present a novel approach to automatic rebuild salient object sequence from video while considering the visual attention rebuilding *** the salient object tracking,salient objects may transfer between frames,which i...
详细信息
We present a novel approach to automatic rebuild salient object sequence from video while considering the visual attention rebuilding *** the salient object tracking,salient objects may transfer between frames,which induces the invalidation of temporal *** formulate the salient object rebuild problem as an energy minimization problem in a conditional random field framework,with salient object features,spatial and temporal coherent *** consider whether the salient object should be rebuilt in three ***,the potential salient object sequence segments are analyzed using the computed salient ***,variables are assigned to indicate whether two pixels from consecutive frames belong to the same salient object ***,dynamic programming algorithm is designed to resolve this energy minimization *** indicate that our approach can deal with the salient object rebuild problem well for automatic visual attention applications and video content analysis.
The dust evolution simulation provides an important way to analyze the impact of dust on the environment. Dynamic data driven application system framework is applied and dust evolution simulation model based on DDDAS ...
详细信息
The dust evolution simulation provides an important way to analyze the impact of dust on the environment. Dynamic data driven application system framework is applied and dust evolution simulation model based on DDDAS framework is proposed. Compared with the simulation model of dust evolution without DDDAS, the functions of the inputs of real-time data and modification of simulation parameters ensure the improvement of the accuracy of dust evolution simulation. The experimental results show that the dust evolution simulation with DDDAS framework is more precise than the simulation without DDDAS, which proves the effectiveness of the dust evolution simulation system based on dynamic data driven.
This paper designs a guide rail for a wall-climbing robot based on passive suction cups. The designed guide rail can guarantee stable climbing of a wall-climbing robot. Firstly, to design the parameters of the guide r...
详细信息
ISBN:
(纸本)9781509037636
This paper designs a guide rail for a wall-climbing robot based on passive suction cups. The designed guide rail can guarantee stable climbing of a wall-climbing robot. Firstly, to design the parameters of the guide rail, properties of the utilized passive suction cup are experimentally studied. Then a guide rail is designed to distribute the forces of the attached suction cups. It guarantees that the front attached suction cup of the robot obtains a large enough reaction force from the attached surface. Experimental results show that the robot prototype with the designed guide rail is able to stably climb a surface without falling off, since the front suction cup of the robot prototype can be completely attached to the climbing surface.
This paper presents and discusses the trajectory tracking control design with friction compensation for a three wheeled omnidirectional mobile robot. Firstly, a dynamic model without considering the friction forces is...
详细信息
ISBN:
(纸本)9781509037636
This paper presents and discusses the trajectory tracking control design with friction compensation for a three wheeled omnidirectional mobile robot. Firstly, a dynamic model without considering the friction forces is derived. Part of the control effort is used to compensate the friction effects, which is estimated by an extended state observer without using a friction model. Then traditional resolved acceleration control is applied to the robot, based on the derived dynamic model. In addition, stability analysis of the controller and observer is presented. Experimental results show that the proposed control design is efficient in compensating the friction effects and in improving the tracking control performances.
Utilizing plane constraints, a linear kinematic calibration algorithm is proposed for industrial robots based on local frame representation of the POE (Product-of-Exponential) formula. Different from conventional cali...
详细信息
ISBN:
(纸本)9781509041039
Utilizing plane constraints, a linear kinematic calibration algorithm is proposed for industrial robots based on local frame representation of the POE (Product-of-Exponential) formula. Different from conventional calibration methods which need expensive measurement equipment, the proposed method only needs a laser range finder mounted onto the robot end-effector and several predefined planes fixed to the base. As it suffices to lump all the kinematic errors into the initial poses of the local coordinate frames, the error model derived from the plane constraints is significantly simplified. An iterative least square algorithm is employed to identify the kinematic errors of the robot. Simulation results demonstrate the effectiveness of the proposed calibration method.
In this paper, we propose output feedback and state feedback controllers for output regulation of linear heterogeneous multi-agent systems. We use a general model for the output regulation problem of linear multi-agen...
详细信息
An obstacle detection method is proposed based on binocular stereo vision for manipulator. Firstly, images are preprocessed after camera calibration such as graying and distortion correction. Then, aiming at the defec...
详细信息
An obstacle detection method is proposed based on binocular stereo vision for manipulator. Firstly, images are preprocessed after camera calibration such as graying and distortion correction. Then, aiming at the defect that feature points are extracted mixed with many edge response points by traditional SIFT(Scale Invariant Feature Transform) algorithm, we propose wavelet transform to extract the edge points, and compare the edge points with feature points to further eliminate unstable feature points. Finally, the effectiveness of the algorithm is verified by obstacle detection and obstacle avoiding experiments. The results demonstrate that the improved SIFT algorithm can remove the edge responses that the DOG operator generates effectively, and improve the stability and anti-noise ability of SIFT algorithm compared with traditional SIFT algorithm.
This paper presents a design scheme of a multiple-fan active control wind tunnel system for outdoor near-surface airflow simulation. The wind tunnel, with a rectangular shape, has 6 swivel plates which can rotate in-9...
详细信息
ISBN:
(纸本)9781467374439
This paper presents a design scheme of a multiple-fan active control wind tunnel system for outdoor near-surface airflow simulation. The wind tunnel, with a rectangular shape, has 6 swivel plates which can rotate in-90°90° at the entrance. Each swivel plate consists of 8 axial flow fans. The stochastic control strategy is used to regulate fan speeds and the angles of swivel plates in order to produce time-varying wind speed and direction. The simulated airflow in the wind tunnel is compared with the outdoor near-surface airflow by analyzing the statistical indicators and multi-scale entropy of wind speed. The experimental results show that the wind tunnel can reliably simulate the outdoor near-surface airflow. This wind tunnel provides a convenient experimental platform for the odor source localization research in the outdoor near-surface breeze condition.
This paper presents the methods and experiment results of three-dimensional gas distribution mapping with a microdrone. With the 3D Kernel DM(kernel extrapolation distribution mapping) algorithm implemented into the m...
详细信息
ISBN:
(纸本)9781467374439
This paper presents the methods and experiment results of three-dimensional gas distribution mapping with a microdrone. With the 3D Kernel DM(kernel extrapolation distribution mapping) algorithm implemented into the micro-drone's onboard computer, the gas distribution map is real-time updated when the micro-drone flies along predefined sweeping path to traverse the search area in outdoor environments. Results show that the concentration grid map obtained can represent the real gas distribution in a certain extent, and demonstrate that gas distribution mapping with a micro-drone is applicable in the realistic scenarios.
暂无评论