The goal of this paper is to develop a unified online motion generation scheme for quadruped lateral-sequence walk and trot gaits based on a linear model predictive control formulation.Specifically,the dynamics of the...The goal of this paper is to develop a unified online motion generation scheme for quadruped lateral-sequence walk and trot gaits based on a linear model predictive control formulation.Specifically,the dynamics of the linear pendulum model is formulated over a predictive horizon by dimensional analysis.Through gait pattern conversion,the lateral-sequence walk and trot gaits of the quadruped can be regarded as unified biped gaits,allowing the dynamics of the linear inverted pendulum model to serve quadruped motion generation.In addition,a simple linearization of the center of pressure constraints for these quadruped gaits is developed for linear model predictive control problem.Furthermore,the motion generation problem can be solved online by quadratic programming with foothold adaptation.It is demonstrated that the proposed unified scheme can generate stable locomotion online for quadruped lateral-sequence walk and trot gaits,both in simulation and on hardware.The results show significant performance improvements compared to previous work.Moreover,the results also suggest the linearly simplified scheme has the ability to robustness against unexpected disturbances.展开更多
For a spherical four-bar linkage,the maximum number of the spherical RR dyad（R：revolute joint）of five-orientation motion generation can be at most 6.However,complete real solution of this problem has seldom been st...For a spherical four-bar linkage,the maximum number of the spherical RR dyad（R：revolute joint）of five-orientation motion generation can be at most 6.However,complete real solution of this problem has seldom been studied.In order to obtain six real RR dyads,based on Strum＇s theorem,the relationships between the design parameters are derived from a 6th-degree univariate polynomial equation that is deduced from the constraint equations of the spherical RR dyad by using Dixon resultant method.Moreover,the Grashof condition and the circuit defect condition are taken into account.Given the relationships between the design parameters and the aforementioned two conditions,two objective functions are constructed and optimized by the adaptive genetic algorithm（AGA）.Two examples with six real spherical RR dyads are obtained by optimization,and the results verify the feasibility of the proposed method.The paper provides a method to synthesize the complete real solution of the five-orientation motion generation,which is also applicable to the problem that deduces to a univariate polynomial equation and requires the generation of as many as real roots.展开更多
The human motion generation model can extract structural features from existing human motion capture data,and the generated data makes animated characters move.The 3D human motion capture sequences contain complex spa...The human motion generation model can extract structural features from existing human motion capture data,and the generated data makes animated characters move.The 3D human motion capture sequences contain complex spatial-temporal structures,and the deep learning model can fully describe the potential semantic structure of human motion.To improve the authenticity of the generated human motion sequences,we propose a multi-task motion generation model that consists of a discriminator and a generator.The discriminator classifies motion sequences into different styles according to their similarity to the mean spatial-temporal templates from motion sequences of 17 crucial human joints in three-freedom degrees.And target motion sequences are created with these styles by the generator.Unlike traditional related works,our model can handle multiple tasks,such as identifying styles and generating data.In addition,by extracting 17 crucial joints from 29 human joints,our model avoids data redundancy and improves the accuracy of model recognition.The experimental results show that the discriminator of the model can effectively recognize diversified movements,and the generated data can correctly fit the actual data.The combination of discriminator and generator solves the problem of low reuse rate of motion data,and the generated motion sequences are more suitable for actual movement.展开更多
Extensive research has explored human motion generation,but the generated sequences are influenced by different motion styles.For instance,the act of walking with joy and sorrow evokes distinct effects on a character...Extensive research has explored human motion generation,but the generated sequences are influenced by different motion styles.For instance,the act of walking with joy and sorrow evokes distinct effects on a character’s motion.Due to the difficulties in motion capture with styles,the available data for style research are also limited.To address the problems,we propose ASMNet,an action and style-conditioned motion generative network.This network ensures that the generated human motion sequences not only comply with the provided action label but also exhibit distinctive stylistic features.To extract motion features from human motion sequences,we design a spatial temporal extractor.Moreover,we use the adaptive instance normalization layer to inject style into the target motion.Our results are comparable to state-of-the-art approaches and display a substantial advantage in both quantitative and qualitative evaluations.The code is available at https://github.com/ZongYingLi/ASMNet.git.展开更多
The correlation between music and human motion has attracted widespread research attention.Although recent studies have successfully generated motion for singers,dancers,and musicians,few have explored motion generati...The correlation between music and human motion has attracted widespread research attention.Although recent studies have successfully generated motion for singers,dancers,and musicians,few have explored motion generation for orchestral conductors.The generation of music-driven conducting motion should consider not only the basic music beats,but also mid-level music structures,high-level music semantic expressions,and hints for different parts of orchestras(strings,woodwind,etc.).However,most existing conducting motion generation methods rely heavily on human-designed rules,which significantly limits the quality of generated motion.Therefore,we propose a novel Music Motion Synchronized Generative Adversarial Network(M^(2)S-GAN),which generates motions according to the automatically learned music representations.More specifically,M^(2)S-GAN is a cross-modal generative network comprising four components:1)a music encoder that encodes the music signal;2)a generator that generates conducting motion from the music codes;3)a motion encoder that encodes the motion;4)a discriminator that differentiates the real and generated motions.These four components respectively imitate four key aspects of human conductors:understanding music,interpreting music,precision and elegance.The music and motion encoders are first jointly trained by a self-supervised contrastive loss,and can thus help to facilitate the music motion synchronization during the following adversarial learning process.To verify the effectiveness of our method,we construct a large-scale dataset,named ConductorMotion100,which consists of unprecedented 100 hours of conducting motion data.Extensive experiments on ConductorMotion100 demonstrate the effectiveness of M^(2)S-GAN.Our proposed approach outperforms various comparison methods both quantitatively and qualitatively.Through visualization,we show that our approach can generate plausible,diverse,and music-synchronized conducting motion.展开更多
Almost all living organisms exhibit autonomic oscillatory activities,which are primarily generated by the rhythmic activities of their neural systems.Several nonlinear oscillator models have been proposed to elucidate...Almost all living organisms exhibit autonomic oscillatory activities,which are primarily generated by the rhythmic activities of their neural systems.Several nonlinear oscillator models have been proposed to elucidate these neural behaviors and subsequently applied to the domain of robot control.However,the oscillation patterns generated by these models are often unpredictable and need to be obtained through parameter search.This study introduces a mathematical model that can be used to analyze multiple neurons connected through fast inhibitory synapses.The characteristic of this oscillator is that its stationary point is stable,but the location of the stationary point changes with the system state.Only through reasonable topology and threshold parameter selection can the oscillation be sustained.This study analyzed the conditions for stable oscillation in two-neuron networks and three-neuron networks,and obtained the basic rules of the phase relationship of the oscillator network established by this model.In addition,this study also introduces synchronization mechanisms into the model to enable it to be synchronized with the sensing pulse.Finally,this study used these theories to establish a robot single leg joint angle generation system.The experimental results showed that the simulated robot could achieve synchronization with human motion,and had better control effects compared to traditional oscillators.展开更多
Inspired by the success of WaveNet in multi-subject speech synthesis,we propose a novel neural network based on causal convolutions for multi-subject motion modeling and generation.The network can capture the intrinsi...Inspired by the success of WaveNet in multi-subject speech synthesis,we propose a novel neural network based on causal convolutions for multi-subject motion modeling and generation.The network can capture the intrinsic characteristics of the motion of different subjects,such as the influence of skeleton scale variation on motion style.Moreover,after fine-tuning the network using a small motion dataset for a novel skeleton that is not included in the training dataset,it is able to synthesize high-quality motions with a personalized style for the novel skeleton.The experimental results demonstrate that our network can model the intrinsic characteristics of motions well and can be applied to various motion modeling and synthesis tasks.展开更多
Human motion modeling is a core technology in computer animation,game development,and humancomputer interaction.In particular,generating natural and coherent in-between motion using only the initial and terminal frame...Human motion modeling is a core technology in computer animation,game development,and humancomputer interaction.In particular,generating natural and coherent in-between motion using only the initial and terminal frames remains a fundamental yet unresolved challenge.Existing methods typically rely on dense keyframe inputs or complex prior structures,making it difficult to balance motion quality and plausibility under conditions such as sparse constraints,long-term dependencies,and diverse motion styles.To address this,we propose a motion generation framework based on a frequency-domain diffusion model,which aims to better model complex motion distributions and enhance generation stability under sparse conditions.Our method maps motion sequences to the frequency domain via the Discrete Cosine Transform(DCT),enabling more effective modeling of low-frequency motion structures while suppressing high-frequency noise.A denoising network based on self-attention is introduced to capture long-range temporal dependencies and improve global structural awareness.Additionally,a multi-objective loss function is employed to jointly optimize motion smoothness,pose diversity,and anatomical consistency,enhancing the realism and physical plausibility of the generated sequences.Comparative experiments on the Human3.6M and LaFAN1 datasets demonstrate that our method outperforms state-of-the-art approaches across multiple performance metrics,showing stronger capabilities in generating intermediate motion frames.This research offers a new perspective and methodology for human motion generation and holds promise for applications in character animation,game development,and virtual interaction.展开更多
基金supported by the National Natural Science Foundation of China(Nos.52305072 and 52122503)Natural Science Foundation of Hebei Province of China(No.E2022203095)+2 种基金University-Industry Collaborative Education Program(No.220603936245709)Cultivation Project for Basic Research and Innovation of Yanshan University(No.2021LGQN004)henzhen Special Fund for Future Industrial Development(No.KJZD20230923114222045).
文摘The goal of this paper is to develop a unified online motion generation scheme for quadruped lateral-sequence walk and trot gaits based on a linear model predictive control formulation.Specifically,the dynamics of the linear pendulum model is formulated over a predictive horizon by dimensional analysis.Through gait pattern conversion,the lateral-sequence walk and trot gaits of the quadruped can be regarded as unified biped gaits,allowing the dynamics of the linear inverted pendulum model to serve quadruped motion generation.In addition,a simple linearization of the center of pressure constraints for these quadruped gaits is developed for linear model predictive control problem.Furthermore,the motion generation problem can be solved online by quadratic programming with foothold adaptation.It is demonstrated that the proposed unified scheme can generate stable locomotion online for quadruped lateral-sequence walk and trot gaits,both in simulation and on hardware.The results show significant performance improvements compared to previous work.Moreover,the results also suggest the linearly simplified scheme has the ability to robustness against unexpected disturbances.
基金Supported by National Natural Science Foundation of China(Grant Nos.51375059,61105103)National Hi-tech Research and Development Program of China(863 Program,Grant No.2011AA040203)Beijing Municipal Natural Science Foundation of China(Grant No.4132032)
文摘For a spherical four-bar linkage,the maximum number of the spherical RR dyad（R：revolute joint）of five-orientation motion generation can be at most 6.However,complete real solution of this problem has seldom been studied.In order to obtain six real RR dyads,based on Strum＇s theorem,the relationships between the design parameters are derived from a 6th-degree univariate polynomial equation that is deduced from the constraint equations of the spherical RR dyad by using Dixon resultant method.Moreover,the Grashof condition and the circuit defect condition are taken into account.Given the relationships between the design parameters and the aforementioned two conditions,two objective functions are constructed and optimized by the adaptive genetic algorithm（AGA）.Two examples with six real spherical RR dyads are obtained by optimization,and the results verify the feasibility of the proposed method.The paper provides a method to synthesize the complete real solution of the five-orientation motion generation,which is also applicable to the problem that deduces to a univariate polynomial equation and requires the generation of as many as real roots.
文摘The human motion generation model can extract structural features from existing human motion capture data,and the generated data makes animated characters move.The 3D human motion capture sequences contain complex spatial-temporal structures,and the deep learning model can fully describe the potential semantic structure of human motion.To improve the authenticity of the generated human motion sequences,we propose a multi-task motion generation model that consists of a discriminator and a generator.The discriminator classifies motion sequences into different styles according to their similarity to the mean spatial-temporal templates from motion sequences of 17 crucial human joints in three-freedom degrees.And target motion sequences are created with these styles by the generator.Unlike traditional related works,our model can handle multiple tasks,such as identifying styles and generating data.In addition,by extracting 17 crucial joints from 29 human joints,our model avoids data redundancy and improves the accuracy of model recognition.The experimental results show that the discriminator of the model can effectively recognize diversified movements,and the generated data can correctly fit the actual data.The combination of discriminator and generator solves the problem of low reuse rate of motion data,and the generated motion sequences are more suitable for actual movement.
基金supported by National Natural Science Foundation of China(No.62203476)Natural Science Foundation of Shenzhen(No.JCYJ20230807120801002).
文摘Extensive research has explored human motion generation,but the generated sequences are influenced by different motion styles.For instance,the act of walking with joy and sorrow evokes distinct effects on a character’s motion.Due to the difficulties in motion capture with styles,the available data for style research are also limited.To address the problems,we propose ASMNet,an action and style-conditioned motion generative network.This network ensures that the generated human motion sequences not only comply with the provided action label but also exhibit distinctive stylistic features.To extract motion features from human motion sequences,we design a spatial temporal extractor.Moreover,we use the adaptive instance normalization layer to inject style into the target motion.Our results are comparable to state-of-the-art approaches and display a substantial advantage in both quantitative and qualitative evaluations.The code is available at https://github.com/ZongYingLi/ASMNet.git.
基金the Natural Science Foundation of Jiangsu Province of China under Grant No.BK20191298the National Natural Science Foundation of China under Grant No.61902110.
文摘The correlation between music and human motion has attracted widespread research attention.Although recent studies have successfully generated motion for singers,dancers,and musicians,few have explored motion generation for orchestral conductors.The generation of music-driven conducting motion should consider not only the basic music beats,but also mid-level music structures,high-level music semantic expressions,and hints for different parts of orchestras(strings,woodwind,etc.).However,most existing conducting motion generation methods rely heavily on human-designed rules,which significantly limits the quality of generated motion.Therefore,we propose a novel Music Motion Synchronized Generative Adversarial Network(M^(2)S-GAN),which generates motions according to the automatically learned music representations.More specifically,M^(2)S-GAN is a cross-modal generative network comprising four components:1)a music encoder that encodes the music signal;2)a generator that generates conducting motion from the music codes;3)a motion encoder that encodes the motion;4)a discriminator that differentiates the real and generated motions.These four components respectively imitate four key aspects of human conductors:understanding music,interpreting music,precision and elegance.The music and motion encoders are first jointly trained by a self-supervised contrastive loss,and can thus help to facilitate the music motion synchronization during the following adversarial learning process.To verify the effectiveness of our method,we construct a large-scale dataset,named ConductorMotion100,which consists of unprecedented 100 hours of conducting motion data.Extensive experiments on ConductorMotion100 demonstrate the effectiveness of M^(2)S-GAN.Our proposed approach outperforms various comparison methods both quantitatively and qualitatively.Through visualization,we show that our approach can generate plausible,diverse,and music-synchronized conducting motion.
基金supported in part by the National Nature Science Foudation under Grant 62333023in part by the Key Research and Development Program of Zhejiang Province under Grant 2021C03050in part by the Scientific Research Project of Agriculture and Social Development of Hangzhou under Grant 20212013B11.
文摘Almost all living organisms exhibit autonomic oscillatory activities,which are primarily generated by the rhythmic activities of their neural systems.Several nonlinear oscillator models have been proposed to elucidate these neural behaviors and subsequently applied to the domain of robot control.However,the oscillation patterns generated by these models are often unpredictable and need to be obtained through parameter search.This study introduces a mathematical model that can be used to analyze multiple neurons connected through fast inhibitory synapses.The characteristic of this oscillator is that its stationary point is stable,but the location of the stationary point changes with the system state.Only through reasonable topology and threshold parameter selection can the oscillation be sustained.This study analyzed the conditions for stable oscillation in two-neuron networks and three-neuron networks,and obtained the basic rules of the phase relationship of the oscillator network established by this model.In addition,this study also introduces synchronization mechanisms into the model to enable it to be synchronized with the sensing pulse.Finally,this study used these theories to establish a robot single leg joint angle generation system.The experimental results showed that the simulated robot could achieve synchronization with human motion,and had better control effects compared to traditional oscillators.
基金We thank the anonymous reviewers for their constructive comments.Weiwei Xu is partially supported by the National Natural Science Foundation of China(No.61732016).
文摘Inspired by the success of WaveNet in multi-subject speech synthesis,we propose a novel neural network based on causal convolutions for multi-subject motion modeling and generation.The network can capture the intrinsic characteristics of the motion of different subjects,such as the influence of skeleton scale variation on motion style.Moreover,after fine-tuning the network using a small motion dataset for a novel skeleton that is not included in the training dataset,it is able to synthesize high-quality motions with a personalized style for the novel skeleton.The experimental results demonstrate that our network can model the intrinsic characteristics of motions well and can be applied to various motion modeling and synthesis tasks.
基金supported by the National Natural Science Foundation of China(Grant No.72161034).
文摘Human motion modeling is a core technology in computer animation,game development,and humancomputer interaction.In particular,generating natural and coherent in-between motion using only the initial and terminal frames remains a fundamental yet unresolved challenge.Existing methods typically rely on dense keyframe inputs or complex prior structures,making it difficult to balance motion quality and plausibility under conditions such as sparse constraints,long-term dependencies,and diverse motion styles.To address this,we propose a motion generation framework based on a frequency-domain diffusion model,which aims to better model complex motion distributions and enhance generation stability under sparse conditions.Our method maps motion sequences to the frequency domain via the Discrete Cosine Transform(DCT),enabling more effective modeling of low-frequency motion structures while suppressing high-frequency noise.A denoising network based on self-attention is introduced to capture long-range temporal dependencies and improve global structural awareness.Additionally,a multi-objective loss function is employed to jointly optimize motion smoothness,pose diversity,and anatomical consistency,enhancing the realism and physical plausibility of the generated sequences.Comparative experiments on the Human3.6M and LaFAN1 datasets demonstrate that our method outperforms state-of-the-art approaches across multiple performance metrics,showing stronger capabilities in generating intermediate motion frames.This research offers a new perspective and methodology for human motion generation and holds promise for applications in character animation,game development,and virtual interaction.
