A humanoid robot may suffer from slip since its two feet are not fixed on the ground. Slip occurrence may induce the loss of robot's balance. The previous literature focused on the detection or estimation of the s...
详细信息
A humanoid robot may suffer from slip since its two feet are not fixed on the ground. Slip occurrence may induce the loss of robot's balance. The previous literature focused on the detection or estimation of the slip occurrence, or only employed the translational acceleration regulation to overcome slip. This study proposes a slip prevention method by coordinating acceleration vector including the horizontal, vertical, and rotational acceleration components. The contribution of this paper is to regulate the translational and rotational acceleration together to prevent the slip occurrence. The effectiveness of our proposed method was verified by simulations and experiments on an actual humanoid robot.
In recent years,deep learning based object detection has achieved great *** methods typically assume that large amount of labeled training data is available,meanwhile,training and test data are independent and identic...
详细信息
In recent years,deep learning based object detection has achieved great *** methods typically assume that large amount of labeled training data is available,meanwhile,training and test data are independent and identically ***,the two assumptions are not always hold in *** many applications,it is prohibitively expensive and time-consuming to obtain large quantities of labeled *** computer graphics technology to generate a large number of labeled data provides a solution to this ***,direct transfer across domains from synthesis to reality often performs poorly due to the presence of domain *** adaptive object detection are concerned with accounting for these types of *** this paper,we present an introduction to these ***,we briefly introduce the object detection and domain ***,the synthetic object detection datasets and related software tools are ***,we present a categorization of approaches,divided into discrepancy-based methods,adversarial discriminative methods,reconstruction-based methods and ***,we also discuss some potential deficiencies of current methods and several open problems which can be explored in future work.
A strictly positive real control problem for delta operator systems in a low frequency range is presented by using the generalized Kalman–Yakubovic˘–Popov lemma. The objective of the strictly positive real control p...
A strictly positive real control problem for delta operator systems in a low frequency range is presented by using the generalized Kalman–Yakubovic˘–Popov lemma. The objective of the strictly positive real control problem is to design a controller such that the transfer function is strictly positive real and the resulting closed-loop system is stable. Sufficient conditions for the low frequency strictly positive real controller of the closed-loop delta operator systems are presented in terms of solutions to a set of linear matrix inequalities. A numerical example is given to illustrate the effectiveness and potential for the developed techniques.
This paper presents an action recognition method based on 2D human body node data in video. This method uses the pose estimation algorithm to detect the human body node data in each frame of video information. We get ...
详细信息
This paper presents an action recognition method based on 2D human body node data in video. This method uses the pose estimation algorithm to detect the human body node data in each frame of video information. We get the two-dimensional coordinates and confidence data of the nodes, and optimize the arrangement of these data into a 3D array form similar to ***, we use the classical two-dimensional convolutional neural network to carry out classification training. The test on UCF-101 data set shows that this method can indeed improve the accuracy of action recognition based on RGB information to a certain extent, and reduce the training cost.
The tracker based on Siamese neural network is currently a technical method with high accuracy in the tracking *** the introduction of transformer in the visual tracking field, the attention mechanism has gradually em...
The tracker based on Siamese neural network is currently a technical method with high accuracy in the tracking *** the introduction of transformer in the visual tracking field, the attention mechanism has gradually emerged in tracking ***, due to the characteristics of attention operation, Transformer usually has slow convergence speed, and its pixel-level correlation discrimination in tracking is more likely to lead to overfitting, which is not conducive to long-term tracking. A brand new framework FAT was designed, which is the improvement of MixFormer. The operation for simultaneous feature extraction and target information integration in MixFormer is retained, and the Mixing block is introduced to suppress the background as much as possible before the information interaction. In addition, a new operation is designed: the result of region-level crosscorrelation is used as a guidance to help the learning of pixel-level cross-correlation in attention, thereby accelerating the model convergence speed and enhancing the model generalization. Finally, a joint loss function is designed to further improve the accuracy of the model. Experiments show that the presented tracker achieves excellent performance on five benchmark datasets.
In this work,a finite-horizon optimal control problem for first-order plus time delay(FOPTD) processes is *** show that if the control horizon is greater than three and the prediction horizon is great than the control...
详细信息
In this work,a finite-horizon optimal control problem for first-order plus time delay(FOPTD) processes is *** show that if the control horizon is greater than three and the prediction horizon is great than the control horizon plus the time delay in discrete time,the optimal controller is not affected by either of the two ***,under these conditions,the controller parameters are explicitly calculated,the closed-loop system is shown to be stable,and the controller is *** problem considered is related to the results on linear quadratic regulation of linear systems with time delays;however,the detailed parameterization of the state-space model introduced by the FOPTD process provides an additional opportunity to investigate the exact controller structure and properties(e.g.,the locations of the closed-loop poles),which are also the major difficulties encountered and overcome in this *** problem is motivated from phenomena experienced in designing industrial model predictive control(MPC) tuning algorithms,and extensive numerical examples indicate that the proposed results speed up the MPC autotuning algorithms by 70%.
intelligentdecision making and efficient trajectory planning are closely related in autonomous driving technology, especially in highway environment full of dynamic interactive traffic participants. This work integra...
详细信息
Twisting between adjacent modules drive is the main way to drive the snake-like robot. It is hard to build a mathematical model for the kind of snake-like robot and the control of it is very complex in the realistic e...
Twisting between adjacent modules drive is the main way to drive the snake-like robot. It is hard to build a mathematical model for the kind of snake-like robot and the control of it is very complex in the realistic environment. In purpose of developing a new kind of snake-like robot which is easy to model, a new way to drive snake-like robot which we call skin drive was invented. With this way, structure of snake-like robot can be very simple and easy to model. Besides control of snake-like robot can be simple. Eventually a snake-like robot was built and it is called round belt drive snake-like robot. Through some experiment and analysis, it is concluded that this new kind of snake-like robot is easy to model and easy to control and its structure is simple to build. In this paper, three important topics were analyzed and elaborated. They are round belt analysis and design, active pulley structure and guide pulley arrangement. Of course, some experiments were done to verify the round belt drive snake-like robot performance. Some experiment process and results were shown at the end in this paper.
Based on the quadruped robot, this paper mainly studies the two directions of the content. The first part mainly introduces the mechanical structure design and the construction of the control system of the quadruped r...
详细信息
Based on the quadruped robot, this paper mainly studies the two directions of the content. The first part mainly introduces the mechanical structure design and the construction of the control system of the quadruped robot, completes the prototype design of the quadruped robot based on hydraulic power system. The second part studies the CPG gait generate method of the quadruped robot based on iterative learning control algorithm. From the principle of bionics, firstly, we use the CPG principle to generate gait, and then use the iterative learning control theory to make the control more optimization.
Recently,unmanned aerial vehicles are widely used in surveillance,aerial photography,power grid line inspections and other *** order to deploy the YOLOv3[1] algorithm on drones,it is necessary to adopt the YOLOv3 algo...
详细信息
Recently,unmanned aerial vehicles are widely used in surveillance,aerial photography,power grid line inspections and other *** order to deploy the YOLOv3[1] algorithm on drones,it is necessary to adopt the YOLOv3 algorithm with fewer parameters and a simpler *** paper implements the model compression of YOLOv3 based on methods such as sparseness,pruning,and knowledge *** paper implements the sparseness of the network by adding L1 regular expressions on the convolutional *** that,redundant channels and layers are removed through channel pruning and layer *** sparse and pruning,the mAP lost a *** using knowledge distillation after pruning,it attempts to recover mAP lost in sparseness and *** this method,the YOLOv3 algorithm can be deployed on embedded platforms such as RK3399 *** evaluate the model on the visdrone2019 *** experimental results show that after model compression,YOLOv3 is more suitable for deployment on embedded platforms.
暂无评论