This paper presents a new approach to the integration of different sensory cues in bubble space for place recognition. In bubble space, bubble surfaces enable the representation of all features in a manner that is imp...
详细信息
In this paper, we focus on requirements analysis of agent based software development and introduce a set of patterns for analysis of (requirements for) learning agents. This paper also proposes a process for producing...
详细信息
This article addresses the model-and data-based event-triggered consensus of heterogeneous leader/follower multi-agent systems(MASs). A dynamic periodic transmission protocol is developed to alleviate the communicatio...
详细信息
This article addresses the model-and data-based event-triggered consensus of heterogeneous leader/follower multi-agent systems(MASs). A dynamic periodic transmission protocol is developed to alleviate the communication and computational burden, where the followers can interact locally with neighbors to approach the dynamics of the leader. Capitalizing on a discrete-time looped-functional, a model-based consensus condition for the closed-loop MASs is derived as linear matrix inequalities(LMIs), along with a design method for obtaining distributed event-triggered controllers and the associated triggering *** collecting noise-corrupted state-input measurements in offline open-loop experiments, a data-based leader/follower MAS representation is derived and employed to address the data-driven consensus control problem without explicit MAS models. This result is subsequently generalized to guarantee an H∞-consensus control performance. Finally, a simulation example is given to corroborate the efficiency of the proposed distributed triggering scheme and the data-driven consensus controller.
In this paper we propose a new multi-view semi-supervised learning algorithm called Local Co-Training (LCT). The proposed algorithm employs a set of local models with vector outputs to model the relations among exampl...
详细信息
As electronic systems are entrusted with increasing numbers of critical tasks, it becomes important that these systems exhibit high levels of reliability. The increasing complexity of modern systems, combined with red...
详细信息
As electronic systems are entrusted with increasing numbers of critical tasks, it becomes important that these systems exhibit high levels of reliability. The increasing complexity of modern systems, combined with reduced development times, makes the production of fault-free systems extremely difficult. Minimising the effects of such faults is a challenging task. Developing a self-repair capability for electronic systems would provide a means to alleviate the effects of a fault. This approach, inspired by healing in biological systems, offers the capability of repairing faults without user intervention. This paper describes a design methodology targeted at self-repairing systems based on modelling systems using the Unified Modelling Language. The approach increases the flexibility of systems by enabling the transfer of functionality between hardware and software. The system configuration is produced by an intelligent reasoning approach, capable of producing optimised configurations at the design stage, or to repair a fault. This design methodology is demonstrated on an example system and the challenges this presents are discussed.
This paper proposes a novel method of combining Rough concepts with Neural Computation. The proposed New Rough Neuron consists of, one Lower Bound Neuron and another Boundary Neuron. The combination is designed in suc...
详细信息
Soccer robot system is a tremendously challenging intelligent system developed to mimic human soccer competition based on the multi discipline research: robotics, intelligentcontrol, computer vision, etc. robot path ...
详细信息
Soccer robot system is a tremendously challenging intelligent system developed to mimic human soccer competition based on the multi discipline research: robotics, intelligentcontrol, computer vision, etc. robot path planning strategy is a very important subject concerning to the performance and intelligence degree of the multi robot system. Therefore, this paper studies the path planning strategy of soccer system by using fuzzy logic. After setting up two fuzziers and two sorts of fuzzy rules for soccer system, fuzzy logic is applied to workspace partition and path revision. The experiment results show that this technique can well enhance the performance and intelligence degree of the system.
Technical analysis is a common method used by financial managers and traders to predict buy and sell trading signals for individual stocks. Unfortunately, it is often the case that each trader, based on their own leve...
详细信息
Technical analysis is a common method used by financial managers and traders to predict buy and sell trading signals for individual stocks. Unfortunately, it is often the case that each trader, based on their own level of expertise, will have a different way of interpreting an indicator, or identifying the time series trend that is currently presented by the stock's price history. This study involves training feed-forward neutral networks to generate buy and sell trading signals. The predictability and profitability results given by the trained neural networks (with both discrete and fuzzified technical indicators) are compared against rule-based models of the technical indicators, as well as a standard benchmark buy-and-hold strategy.
The paper gives a comprehensive description about how to construct a web server for remote measuring and control needs, proposes a solution under Linux software and HMS30C7202 embedded hardware platform. Combined with...
详细信息
Proposes an adaptive learning method based on reinforcement learning for robot path planning problem, which enables the robot to adaptively learn and perform effective path planning, to avoid the moving obstacles and ...
详细信息
Proposes an adaptive learning method based on reinforcement learning for robot path planning problem, which enables the robot to adaptively learn and perform effective path planning, to avoid the moving obstacles and reach the target. Thereby achieving automatic construction of path planning strategy and making the system adaptive to multi robots system dynamic environments, and concludes from computer simulation experiment that the method is powerful to solve the problem of multi robot path planning, and it is a meaningful try to apply reinforcement learning techniques in multi robot systems to develop the system’s intelligence degree.
暂无评论