Motion infilling is a fundamental and challenging research field in human motion modeling and analysis, which aims to generate natural and visually coherent transitions to fill in missing motion frames based on the st...
详细信息
ISBN:
(数字)9798350340266
ISBN:
(纸本)9798350340273
Motion infilling is a fundamental and challenging research field in human motion modeling and analysis, which aims to generate natural and visually coherent transitions to fill in missing motion frames based on the start and end motion sequences. However, most current methods ignore the spatial structure formed by joints, which may lose some spatial information. This work proposes a novel spatiotemporal graph U-Net that supports flexible inputs for skeleton-based motion infilling. We apply spatiotemporal graph convolutional layers, skeleton pooling layers, and skeleton unpooling layers to extract spatial and temporal features in the motion sequence. At the same time, we use the U-Net structure to integrate the information in the start and end motion sequences. In addition, the generative adversarial mechanism is introduced to ensure the generated skeleton poses are smooth and natural. We conduct experiments on two motion datasets, including one large-scale public dataset and one self-built dataset. The model inputs are joint quaternions or joint coordinates. Experimental results show that our method can improve the performance of skeleton-based motion infilling and achieve state-of-the-art results when using joint coordinates as model input.
Robotics and haptic systems have allowed new and diverse applications in the field of medicine, such as assisted surgery and teleoperation which have increasingly stringent requirements for accuracy, convergence, and ...
详细信息
ISBN:
(数字)9798350393965
ISBN:
(纸本)9798350393972
Robotics and haptic systems have allowed new and diverse applications in the field of medicine, such as assisted surgery and teleoperation which have increasingly stringent requirements for accuracy, convergence, and low computational consumption. In this paper an adaptive PID control law (Proportional Integral Derivative controller, PID), of indirect architecture is presented for movement paths in a haptic system of open chain, where the identification of the plant is through a quaternionic wavelet neural network (Quaternion Wavelet Neural Network, QWNN) for tune the PID values, this allows the optimal movement into the regions of the workspace.
This paper is devoted to the attitude tracking problem of a near-ground quadrotor, where the aerodynamic force of propellers is impaired under the ground effect. An altitude-dependent dynamic model for attitude contro...
详细信息
ISBN:
(数字)9798350387780
ISBN:
(纸本)9798350387797
This paper is devoted to the attitude tracking problem of a near-ground quadrotor, where the aerodynamic force of propellers is impaired under the ground effect. An altitude-dependent dynamic model for attitude control of the quadrotor is formulated with consideration of actuator dynamics, upon which an attitude tracking system for near-ground quadrotor is obtained, instead of only considering the altitude control problem via altitude-dependent total lift force in the past studies. By selecting the altitude as a scheduling variable, the attitude tracking control system is further described via a switched linear parameter varying scheme, which has not been considered before. The stability condition and the H
∞
performance are obtained, then an altitude-dependent attitude controller is designed, where the effectiveness and advantages are demonstrated via simulations.
This paper investigates the issue of flight control of fixed-wing aircraft (FwA) in heavy payload airdrop tasks, where smooth control commands are considered for actuators. The case that heavy payloads are mounted out...
详细信息
ISBN:
(数字)9798350387780
ISBN:
(纸本)9798350387797
This paper investigates the issue of flight control of fixed-wing aircraft (FwA) in heavy payload airdrop tasks, where smooth control commands are considered for actuators. The case that heavy payloads are mounted out of the symmetry plane is taken into account for dynamic modeling of the FwA, where the dynamic model is split into different modes corresponding to various situations of mounted payloads. Considering the dramatic variations of mass and center of gravity, a multi-equilibria switched system (MESS) is formulated for flight control, where the stable (unstable) modes correspond to the cases of symmetry (unsymmetry) mounting to ensure that the airdrop is with small rolling/yawing angles. By inserting the smooth portion, a smooth switched controller and the corresponding dwell time condition is obtained for MESS to tackle the flight control problem in heavy payload airdrop of FwA, which has not been considered before. The effectiveness and advantages of the proposed controller are demonstrated via simulations.
Paraphrase identification, with the objective to determine whether two sentences are paraphrases of each other, has fostered many applications including natural language inference and document retrieval. Traditional m...
详细信息
ISBN:
(数字)9798350354560
ISBN:
(纸本)9798350354577
Paraphrase identification, with the objective to determine whether two sentences are paraphrases of each other, has fostered many applications including natural language inference and document retrieval. Traditional machine learning models for paraphrase identification are generally based on hand-crafted features. Recent deep neural network models generally take semantic vectors as inputs and generate classification results directly. However, existing solutions still have much room for improvement. First, existing models do not fully exploit sequence information (i.e., the order of words) and topic information, which are important for text under-standing. Second, although deep models achieve promising results, they require a large amount of labelled data and have low interpretability. To cope with these problems, we propose a new model called PIM-ST (Paraphrase Identification Model Incorporating Sequence and Topic Information). Specifically, besides semantic information, PIM-ST extracts sequence information and topic information, and adopts a simple machine learning model LightGBM as the classifier. We conduct comparative experiments and ablation studies on two public datasets Microsoft Research Paraphrase Corpus (MSRP) and Quora Question Pairs. Experimental results show that PIM-ST achieves better results than existing state-of-the-art models.
This study presents an adaptive fuzzy broad-learning neural control (AFBNC) system applied to a reaction wheel pendulum without knowing its dynamic model. The AFBNC system uses a fuzzy broad-learning system (FBLS) to ...
详细信息
SOME/IP is a communication protocol introduced in automotive industry to enable Service Oriented Architecture which is not possible using traditional automotive protocols such as CAN or FlexRay. A trend emerging in th...
详细信息
This work incorporates an adaptive learning-based boosting (ADboost) ML classifier to classify four types of fuel: agricultural residue, coals, wood, and produced biomass. Further, the ADboost’s hyperparameters, such...
详细信息
ISBN:
(数字)9798350340204
ISBN:
(纸本)9798350340211
This work incorporates an adaptive learning-based boosting (ADboost) ML classifier to classify four types of fuel: agricultural residue, coals, wood, and produced biomass. Further, the ADboost’s hyperparameters, such as learning rate, maximum number of splits, and minimum leaf size are adjusted using teaching learning-based optimization (TLBO), resulting in TADboost. The performance of TADboost is compared against various popular ML (NN, BAG, NB, and SVM) models. Simulated result reveals that the suggested classifiers outperform other compared ML classifiers for fuel classification with the classification accuracy, precision, recall, F1-score, and kappa as 0.9659, 0.9671, 0.9449, 0.9558, and 0.9482, respectively.
In this work, we present the modeling of the dynamics of a robot manipulator using the Newton-Euler algorithm in the conformal algebra framework. The modeling of the dynamics of robot manipulators is currently done us...
详细信息
ISBN:
(数字)9798350362343
ISBN:
(纸本)9798350362350
In this work, we present the modeling of the dynamics of a robot manipulator using the Newton-Euler algorithm in the conformal algebra framework. The modeling of the dynamics of robot manipulators is currently done using the Euler-Lagrange formulation which is a batch type of computation. In contrast, in this paper, we propose a recursive algorithm for the modeling of the dynamics of robot manipulators using the Newton-Euler algorithm in the conformal geometric algebra framework.
Dataflow architecture is a promising parallel computing platform that provides high performance, high efficiency and flexibility. Since the execution of a loop is usually the most time-consuming in a program, extracti...
详细信息
暂无评论