RoboCup is a challenging research domain, in which the full integration of AI and intelligent robot fields are involved. A new autonomous mobile robot named RoboNaut is introduced which satisfies the standard of RoboC...
详细信息
RoboCup is a challenging research domain, in which the full integration of AI and intelligent robot fields are involved. A new autonomous mobile robot named RoboNaut is introduced which satisfies the standard of RoboCup middle sized league. Firstly, its structure is discussed, which consists of mechanical layer, driving layer, and calculating layer. Such an architecture guarantees a robust and concise implementation. Then each layer is discussed in detail. Step motors are used as actuators in the mechanical layer. Driving of motors and sensors are achieved through a mixed signal processor in the driving layer. The image processing, wireless communicating and decision making are done on a CIII800 processor in the calculating layer. Owning to a satisfied performance-price ratio, the RoboNaut is recommended as an open platform for education and research.
The accuracy of concrete strength inspection has a great influence on the safety evaluation of the building. In order to increase the accuracy, Fuzzy Neural Network (FNN) was built up to evaluate concrete stmngth: It ...
详细信息
ISBN:
(纸本)0780382730
The accuracy of concrete strength inspection has a great influence on the safety evaluation of the building. In order to increase the accuracy, Fuzzy Neural Network (FNN) was built up to evaluate concrete stmngth: It takes full advantage of the characteristics of the common concrete testing methods: drill and rebound, and the abilities of FNN including automatic learning, generation and fuzzy logic inference. The experiment shows that the max relative error of the predicted results is 1.12%, which is satisfied with the requirements of the engineering. The method effieieatly maps the complex non-linear relationship between the drill values and the rebound values, and provides a efficient way for the concrete strength inspection and evaluation.
Intrusion detection is an important technique in the defense-in-depth network security framework and a hot topic in computer network security in recent years. In this paper, a new efficient intrusion detection method ...
详细信息
ISBN:
(纸本)0780384032
Intrusion detection is an important technique in the defense-in-depth network security framework and a hot topic in computer network security in recent years. In this paper, a new efficient intrusion detection method based on hidden Markov models (HMMs) is presented. HMMs are applied to model the normal program behaviors using traces of system calls issued by processes. The output probability of a sequence of system calls is calculated by the normal model built. If the probability of a sequence in a trace is below a certain threshold, the sequence is flagged as a mismatch. If the ratio between the mismatches and all the sequences in a trace exceeds another threshold, the trace is then considered as a possible intrusion. The method is implemented and tested on the sendmail system call data from the University of New Mexico. Experimental results show that the performance of the proposed method in intrusion detection is better than other methods.
Q-learning is a typical kind of intelligent learning method for machines. As the scale of state space and action space get larger, the speed of learning convergence will slow down and the learning time will grow up. T...
详细信息
Q-learning is a typical kind of intelligent learning method for machines. As the scale of state space and action space get larger, the speed of learning convergence will slow down and the learning time will grow up. To solve this problem, Q-learning with fuzzy priori knowledge is presented. By using fuzzy integrated decision-making to process expert knowledge and environmental information, this method optimizes initial states of Q-learning, equips it with a better learning foundation and improves its learning efficiency and convergence speed. Finally, the application of the method to the robot soccer system and the simulation result shows the feasibility and the validity of it.
A modification of evolutionary programming or evolution strategies for n-dimensional global optimization is proposed. Based on the ergodicity and inherent-randomness of chaos, the main characteristic of the new algori...
详细信息
A modification of evolutionary programming or evolution strategies for n-dimensional global optimization is proposed. Based on the ergodicity and inherent-randomness of chaos, the main characteristic of the new algorithm which includes two phases is that chaotic behavior is exploited to conduct a rough search of the problem space in order to find the promising individuals in Phase Ⅰ. Adjustment strategy of step-length and intensive searches in Phase Ⅱ are *** population sequences generated by the algorithm asymptotically converge to global optimal solutions with probability one. The proposed algorithm is applied to several typical test problems. Numerical results illustrate that this algorithm can more efficiently solve complex global optimization problems than evolutionary programming and evolution strategies in most cases.
In this paper the interconnection problem in signal-flow digital simulation models is raised. A type of middleware, called virtual electrical bus model, is proposed to solve such a problem in electric vehicle powertra...
详细信息
ISBN:
(纸本)0780379543
In this paper the interconnection problem in signal-flow digital simulation models is raised. A type of middleware, called virtual electrical bus model, is proposed to solve such a problem in electric vehicle powertrain, thus extends the scalability of the system. Analysis on the process of digital simulation shows that models applying the middleware can be compatible to traditional ones. The model is implemented in the Matlab/Simulink environment. The stability of the middleware is studied and some experimental results in the modeling of electrical vehicles powertrain is given.
In this paper we present a speech recognition and verification method based on the integration of likelihood and likelihood ratio. Speech recognition and verification is unified in one-phase framework. A modified aggl...
详细信息
In this paper we present a speech recognition and verification method based on the integration of likelihood and likelihood ratio. Speech recognition and verification is unified in one-phase framework. A modified agglomerative hierarchical clustering algorithm is adopted to train the alternative model used in speech verification. In the process of decoding likelihood ratio is combined with likelihood to get the combination score for searching the final results. Our experimental results showed that false-alarm rate get decreased a lot with only slight loss in accuracy rate.
By generalizing the learning rate parameter to a learning rate matrix, this paper proposes a grading learning algorithm for blind source separation. The whole learning process is divided into three stages: initial sta...
详细信息
By generalizing the learning rate parameter to a learning rate matrix, this paper proposes a grading learning algorithm for blind source separation. The whole learning process is divided into three stages: initial stage, capturing stage and tracking stage. In different stages, different learning rates are used for each output component, which is determined by its dependency on other output components. It is shown that the grading learning algorithm is equivariant and can keep the separating matrix from becoming singular. Simulations show that the proposed algorithm can achieve faster convergence, better steady-state performance and higher numerical robustness, as compared with the existing algorithms using fixed, time-descending and adaptive learning rates.
Multi-Agent System (MAS) has been a wide used and effective method to solve distributed AI problems. In this paper, we simplify the biological mechanism in tree bark growth and build a MAS model to simulate the genera...
详细信息
暂无评论