With the continuous deepening of Artificial Neural Network(ANN)research,ANN model structure and function are improving towards diversification and intelligence.However,the model is more evaluated from the pros and con...With the continuous deepening of Artificial Neural Network(ANN)research,ANN model structure and function are improving towards diversification and intelligence.However,the model is more evaluated from the pros and cons of the problem-solving results and the lack of evaluation from the biomimetic aspect of imitating neural networks is not inclusive enough.Hence,a new ANN models evaluation strategy is proposed from the perspective of bionics in response to this problem in the paper.Firstly,four classical neural network models are illustrated:Back Propagation(BP)network,Deep Belief Network(DBN),LeNet5 network,and olfactory bionic model(KIII model),and the neuron transmission mode and equation,network structure,and weight updating principle of the models are analyzed qualitatively.The analysis results show that the KIII model comes closer to the actual biological nervous system compared with other models,and the LeNet5 network simulates the nervous system in depth.Secondly,evaluation indexes of ANN are constructed from the perspective of bionics in this paper:small-world,synchronous,and chaotic characteristics.Finally,the network model is quantitatively analyzed by evaluation indexes from the perspective of bionics.The experimental results show that the DBN network,LeNet5 network,and BP network have synchronous characteristics.And the DBN network and LeNet5 network have certain chaotic characteristics,but there is still a certain distance between the three classical neural networks and actual biological neural networks.The KIII model has certain small-world characteristics in structure,and its network also exhibits synchronization characteristics and chaotic characteristics.Compared with the DBN network,LeNet5 network,and the BP network,the KIII model is closer to the real biological neural network.展开更多
The landing buffer is an important problem in the research on bionic locust jumping robots, and the different modes of landing and buffering can affect the dynamic performance of the buffering process significantly. B...The landing buffer is an important problem in the research on bionic locust jumping robots, and the different modes of landing and buffering can affect the dynamic performance of the buffering process significantly. Based on an experimental observation, the different modes of landing and buffering are determined, which include the different numbers of landing legs and different motion modes of legs in the buffering process. Then a bionic locust mechanism is established, and the springs are used to replace the leg muscles to achieve a buffering effect. To reveal the dynamic performance in the buffering process of the bionic locust mechanism, a dynamic model is established with different modes of landing and buffering. In particular, to analyze the buffering process conveniently, an equivalent vibration dynamic model of the bionic locust mechanism is proposed.Given the support forces of the ground to the leg links, which can be obtained from the dynamic model, the spring forces of the legs and the impact resistance of each leg are the important parameters affecting buffering performance, and evaluation principles for buffering performance are proposed according to the aforementioned parameters. Based on the dynamic model and these evaluation principles, the buffering performances are analyzed and compared in different modes of landing and buffering on a horizontal plane and an inclined plane. The results show that the mechanism with the ends of the legs sliding can obtain a better dynamic performance. This study offers primary theories for buffering dynamics and an evaluation of landing buffer performance,and it establishes a theoretical basis for studies and engineering applications.展开更多
In this paper, a novel bionic model and its performance in pattern recognition are presented and discussed. The model is constructed from a bulb model and a three-layered cortical model, mimicking the main features of...In this paper, a novel bionic model and its performance in pattern recognition are presented and discussed. The model is constructed from a bulb model and a three-layered cortical model, mimicking the main features of the olfactory system. The olfactory bulb and cortex models are connected by feedforward and feedback fibers with distributed delays. The Breast Cancer Wisconsin dataset consisting of data from 683 patients divided into benign and malignant classes is used to demonstrate the capacity of the model to learn and recognize patterns, even when these are deformed versions of the originally learned patterns. The performance of the novel model was compared with three artificial neural networks (ANNs), a back-propagation network, a support vector machine classifier, and a radial basis function classifier. All the ANNs and the olfactory bionic model were tested in a benchmark study of a standard dataset. Experimental results show that the bionic olfactory system model can learn and classify patterns based on a small training set and a few learning trials to reflect biological intelligence to some extent.展开更多
In this paper, a wild boar head was taken as the bionic research object for the development of new ridgers, a kind of plough. The reverse engineering technology was adopted to obtain the surface geometrical informatio...In this paper, a wild boar head was taken as the bionic research object for the development of new ridgers, a kind of plough. The reverse engineering technology was adopted to obtain the surface geometrical information of the head. Several three-dimensional （3D） point clouds of the head were captured first using a non-touch laser scanner, and an integrated point cloud was generated by aligning these point clouds using UG/Imageware. Then, the digital surface model of the head was rebuilt by means of CATIA. The characteristic curves of the surface model were analyzed. The results show that the average error between the rebuilt surface and the point cloud is -0.431 ram. The max curvature of the ridge on the neb of the head is 0.187 mm^-1, and the max and rain Gauss curvatures on the surface are 0.008 mm^-2 and -0.002 mm^-2. These geometrical information are the essential parameters for biomimetics study of the ridger.展开更多
In order to realize the small error attitude transformation of a free floating space robot,a new method of three degrees of freedom（ DOF） attitude transformation was proposed for the space robot using a bionic joint...In order to realize the small error attitude transformation of a free floating space robot,a new method of three degrees of freedom（ DOF） attitude transformation was proposed for the space robot using a bionic joint. A general kinematic model of the space robot was established based on the law of linear and angular momentum conservation. A combinational joint model was established combined with bionic joint and closed motion. The attitude transformation of planar,two DOF and three DOF is analyzed and simulated by the model,and it is verified that the feasibility of attitude transformation in three DOF space. Finally,the specific scheme of disturbance elimination in attitude transformation is presented and simulation results are obtained.Therefore,the range of application field of the bionic joint model has been expanded.展开更多
The optimisation of earthquake resistance of high buildings by multi-tuned mass dampers was investigated using bionic algorithms. In bionic or evolutionary optimisation studies the properties of parents are crossed an...The optimisation of earthquake resistance of high buildings by multi-tuned mass dampers was investigated using bionic algorithms. In bionic or evolutionary optimisation studies the properties of parents are crossed and mutated to produce a new generation with slightly different properties. The kids which best satisfy the object of the study, become the parents of the next generation. After a series of generations essential improvements of the object may be observed. Tuned mass dampers are widely used to reduce the impact of dynamic excitations on structures. A single mass system and multi-mass oscillators help to explain the mechanics of the dampers. To apply the bionic optimisation strategy to high buildings with passive tuned mass dampers subject to seismic loading a special beam element has been developed. In addition to the 6 degrees of freedom of a conventional beam element, it has 2 degrees of freedom for the displacements of the dampers. It allows for fast studies of many variants. As central result, efficient designs for damping systems along the height of an edifice are found. The impact on the structure may be reduced essentially by the use of such dampers designed to interact in an optimal way.展开更多
In dry attachment systems of spiders and geckos, van der Waals forces mediate attraction between substrate and animal tarsus. In particular, the scopula of Evarcha arcuata spiders allows for reversible attachment and ...In dry attachment systems of spiders and geckos, van der Waals forces mediate attraction between substrate and animal tarsus. In particular, the scopula of Evarcha arcuata spiders allows for reversible attachment and easy detachment to a broad range of surfaces. Hence, reproducing the scopula's roughness compatibility while maintaining anti-bunching features and dirt particle repellence behavior is a central task for a biomimetic transfer to an engineered model. In the present work we model the scopula of E. arcuata from a mechano-elastic point of view analyzing the influence of its hierarchical structure on the attachment behavior. By considering biological data of the gecko and spider, and the simulation results, the adhesive capabilities of the two animals are compared and important confirmations and new directives in order to reproduce the overall structure are found. Moreover, a possible suggestion of how the spider detaches in an easy and fast manner is proposed and supported by the results.展开更多
Conventional moving target detection focuses on algorithms to improve detection efficiency. These algorithms pay less attention to the image acquisition means, and usually solve specific problems. This often results i...Conventional moving target detection focuses on algorithms to improve detection efficiency. These algorithms pay less attention to the image acquisition means, and usually solve specific problems. This often results in poor flexibility and reus- ability. Insect compound eyes offer unique advantages for moving target detection and these advantages have attracted the attention of many researchers in recent years. In this paper we proposed a new system for moving target detection. We used the detection mechanism of insect compound eyes for the simulation of the characteristics of structure, control, and function. We discussed the design scheme of the system, the development of the bionic control circuit, and introduced the proposed mathe- matical model of bionic cqmpound eyes for data acquisition and object detection. After this the integrated system was described and discussed. Our paper presents a novel approach for moving target detection. This approach effectively tackles some of the well-known problems in the field of view, resolution, and real-time processing problems in moving target detection.展开更多
基金supported by the Open Research Project of the State Key Laboratory of Industrial Control Technology(No.ICT2021B10)the Natural Science Foundation of Hunan Province(2021JJ30456)+2 种基金the Open Fund of Science and Technology on Parallel and Distributed Processing Laboratory(WDZC20205500119)the Hunan Provincial Science and Technology Department High-tech Industry Science and Technology Innovation Leading Project(2020GK2009)the Scientific and Technological Progress and Innovation Program of the Transportation Department of Hunan Province(201927),etc.
文摘With the continuous deepening of Artificial Neural Network(ANN)research,ANN model structure and function are improving towards diversification and intelligence.However,the model is more evaluated from the pros and cons of the problem-solving results and the lack of evaluation from the biomimetic aspect of imitating neural networks is not inclusive enough.Hence,a new ANN models evaluation strategy is proposed from the perspective of bionics in response to this problem in the paper.Firstly,four classical neural network models are illustrated:Back Propagation(BP)network,Deep Belief Network(DBN),LeNet5 network,and olfactory bionic model(KIII model),and the neuron transmission mode and equation,network structure,and weight updating principle of the models are analyzed qualitatively.The analysis results show that the KIII model comes closer to the actual biological nervous system compared with other models,and the LeNet5 network simulates the nervous system in depth.Secondly,evaluation indexes of ANN are constructed from the perspective of bionics in this paper:small-world,synchronous,and chaotic characteristics.Finally,the network model is quantitatively analyzed by evaluation indexes from the perspective of bionics.The experimental results show that the DBN network,LeNet5 network,and BP network have synchronous characteristics.And the DBN network and LeNet5 network have certain chaotic characteristics,but there is still a certain distance between the three classical neural networks and actual biological neural networks.The KIII model has certain small-world characteristics in structure,and its network also exhibits synchronization characteristics and chaotic characteristics.Compared with the DBN network,LeNet5 network,and the BP network,the KIII model is closer to the real biological neural network.
基金supported by the National Natural Science Foundation of China (Grant 51375035)the Specialized Research Fund for the Doctoral Program of Higher Education of China (Grant 20121102110021)
文摘The landing buffer is an important problem in the research on bionic locust jumping robots, and the different modes of landing and buffering can affect the dynamic performance of the buffering process significantly. Based on an experimental observation, the different modes of landing and buffering are determined, which include the different numbers of landing legs and different motion modes of legs in the buffering process. Then a bionic locust mechanism is established, and the springs are used to replace the leg muscles to achieve a buffering effect. To reveal the dynamic performance in the buffering process of the bionic locust mechanism, a dynamic model is established with different modes of landing and buffering. In particular, to analyze the buffering process conveniently, an equivalent vibration dynamic model of the bionic locust mechanism is proposed.Given the support forces of the ground to the leg links, which can be obtained from the dynamic model, the spring forces of the legs and the impact resistance of each leg are the important parameters affecting buffering performance, and evaluation principles for buffering performance are proposed according to the aforementioned parameters. Based on the dynamic model and these evaluation principles, the buffering performances are analyzed and compared in different modes of landing and buffering on a horizontal plane and an inclined plane. The results show that the mechanism with the ends of the legs sliding can obtain a better dynamic performance. This study offers primary theories for buffering dynamics and an evaluation of landing buffer performance,and it establishes a theoretical basis for studies and engineering applications.
基金Project supported by the National Natural Science Foundation of China (Nos. 60874098 and 60911130129)the High-Tech Research and Development Program (863) of China (No. 2007AA042103)+1 种基金the National Creative Research Groups Science Foundation of China (No. 60721062)the Project of Introducing Talents for Chinese University Disciplinal Innovation (111 Project, No. B07031)
文摘In this paper, a novel bionic model and its performance in pattern recognition are presented and discussed. The model is constructed from a bulb model and a three-layered cortical model, mimicking the main features of the olfactory system. The olfactory bulb and cortex models are connected by feedforward and feedback fibers with distributed delays. The Breast Cancer Wisconsin dataset consisting of data from 683 patients divided into benign and malignant classes is used to demonstrate the capacity of the model to learn and recognize patterns, even when these are deformed versions of the originally learned patterns. The performance of the novel model was compared with three artificial neural networks (ANNs), a back-propagation network, a support vector machine classifier, and a radial basis function classifier. All the ANNs and the olfactory bionic model were tested in a benchmark study of a standard dataset. Experimental results show that the bionic olfactory system model can learn and classify patterns based on a small training set and a few learning trials to reflect biological intelligence to some extent.
基金The authors are grateful for the financial support by the National Natural Science Foundation of China(Grant No.50635030)the Scientific and Technological Development Project of Jilin Province(Grant No.20050539)the"985 Project"of Jilin University.
文摘In this paper, a wild boar head was taken as the bionic research object for the development of new ridgers, a kind of plough. The reverse engineering technology was adopted to obtain the surface geometrical information of the head. Several three-dimensional （3D） point clouds of the head were captured first using a non-touch laser scanner, and an integrated point cloud was generated by aligning these point clouds using UG/Imageware. Then, the digital surface model of the head was rebuilt by means of CATIA. The characteristic curves of the surface model were analyzed. The results show that the average error between the rebuilt surface and the point cloud is -0.431 ram. The max curvature of the ridge on the neb of the head is 0.187 mm^-1, and the max and rain Gauss curvatures on the surface are 0.008 mm^-2 and -0.002 mm^-2. These geometrical information are the essential parameters for biomimetics study of the ridger.
文摘In order to realize the small error attitude transformation of a free floating space robot,a new method of three degrees of freedom（ DOF） attitude transformation was proposed for the space robot using a bionic joint. A general kinematic model of the space robot was established based on the law of linear and angular momentum conservation. A combinational joint model was established combined with bionic joint and closed motion. The attitude transformation of planar,two DOF and three DOF is analyzed and simulated by the model,and it is verified that the feasibility of attitude transformation in three DOF space. Finally,the specific scheme of disturbance elimination in attitude transformation is presented and simulation results are obtained.Therefore,the range of application field of the bionic joint model has been expanded.
文摘The optimisation of earthquake resistance of high buildings by multi-tuned mass dampers was investigated using bionic algorithms. In bionic or evolutionary optimisation studies the properties of parents are crossed and mutated to produce a new generation with slightly different properties. The kids which best satisfy the object of the study, become the parents of the next generation. After a series of generations essential improvements of the object may be observed. Tuned mass dampers are widely used to reduce the impact of dynamic excitations on structures. A single mass system and multi-mass oscillators help to explain the mechanics of the dampers. To apply the bionic optimisation strategy to high buildings with passive tuned mass dampers subject to seismic loading a special beam element has been developed. In addition to the 6 degrees of freedom of a conventional beam element, it has 2 degrees of freedom for the displacements of the dampers. It allows for fast studies of many variants. As central result, efficient designs for damping systems along the height of an edifice are found. The impact on the structure may be reduced essentially by the use of such dampers designed to interact in an optimal way.
基金supported by the ARIADNA scheme (study ID 06/6201) of the European Space Agency
文摘In dry attachment systems of spiders and geckos, van der Waals forces mediate attraction between substrate and animal tarsus. In particular, the scopula of Evarcha arcuata spiders allows for reversible attachment and easy detachment to a broad range of surfaces. Hence, reproducing the scopula's roughness compatibility while maintaining anti-bunching features and dirt particle repellence behavior is a central task for a biomimetic transfer to an engineered model. In the present work we model the scopula of E. arcuata from a mechano-elastic point of view analyzing the influence of its hierarchical structure on the attachment behavior. By considering biological data of the gecko and spider, and the simulation results, the adhesive capabilities of the two animals are compared and important confirmations and new directives in order to reproduce the overall structure are found. Moreover, a possible suggestion of how the spider detaches in an easy and fast manner is proposed and supported by the results.
基金The work presented in this paper is supported by the Scholarship for International Young Scientists of NSFC （National Natural Science Foundation of China） （1D： 41050110441）.
文摘Conventional moving target detection focuses on algorithms to improve detection efficiency. These algorithms pay less attention to the image acquisition means, and usually solve specific problems. This often results in poor flexibility and reus- ability. Insect compound eyes offer unique advantages for moving target detection and these advantages have attracted the attention of many researchers in recent years. In this paper we proposed a new system for moving target detection. We used the detection mechanism of insect compound eyes for the simulation of the characteristics of structure, control, and function. We discussed the design scheme of the system, the development of the bionic control circuit, and introduced the proposed mathe- matical model of bionic cqmpound eyes for data acquisition and object detection. After this the integrated system was described and discussed. Our paper presents a novel approach for moving target detection. This approach effectively tackles some of the well-known problems in the field of view, resolution, and real-time processing problems in moving target detection.
