With the development of artificial intelligence, communication, computer and other related technologies, it becomes feasible to rebuild traditional railway with such advanced technologies in order to establish a new g...With the development of artificial intelligence, communication, computer and other related technologies, it becomes feasible to rebuild traditional railway with such advanced technologies in order to establish a new generation railway transport system. The railway intelligent transportation system is the trend of railway transportation system in China, and it is also the research focus of international railway transport industry. This paper presents the definition, characters, architecture, key technologies and developing pattern of the RITS(railway intelligent transportation system). Then three typical applications are introduced. Finally, the prospect of the RITS is summarized.展开更多
The research studied the influences of high temperature, high pressure, high humidity, noise and other harmful factors in mining conditions on the people health and safety, and investigated the impacts of confined env...The research studied the influences of high temperature, high pressure, high humidity, noise and other harmful factors in mining conditions on the people health and safety, and investigated the impacts of confined environmental on human physiology factors, including temperature, humidity, noise, pressure,toxic and harmful gases in terms of environmental characteristics in underground mines and an artificial intelligence system for simulation of the environment in a confined space of deep mines. Our results show that the systolic pressure, diastolic pressure, mean pressure, heart rate, respiratory rate, typing test speed and memory level percentage are negatively correlated with temperature value, and positively correlated with humidity value; the human temperature and weight are positively correlated with temperature value, and negatively correlated with humidity value. This research lays the foundation for the study of interaction between the deep confined space environment and safety behavior.展开更多
This paper will discuss the computerised developm en t control and approval system being developed for the Planning and Development C ontrol Department, City Hall of Kuala Lumpur, with stress on the GIS architectur e ...This paper will discuss the computerised developm en t control and approval system being developed for the Planning and Development C ontrol Department, City Hall of Kuala Lumpur, with stress on the GIS architectur e developed within the system. The prospects and challenges towards implementati on of the system are also discussed.展开更多
The application of reinforcement learning is widely used by multi-agent systems in recent years. An agent uses a multi-agent system to cooperate with other agents to accomplish the given task, and one agent′s behavio...The application of reinforcement learning is widely used by multi-agent systems in recent years. An agent uses a multi-agent system to cooperate with other agents to accomplish the given task, and one agent′s behavior usually affects the others′ behaviors. In traditional reinforcement learning, one agent takes the others location, so it is difficult to consider the others′ behavior, which decreases the learning efficiency. This paper proposes multi-agent reinforcement learning with cooperation based on eligibility traces, i.e. one agent estimates the other agent′s behavior with the other agent′s eligibility traces. The results of this simulation prove the validity of the proposed learning method.展开更多
基金funded by the National Natural Science Foundation of China ( No. 61074151) Nation Science and Technology Support Program ( No. T1DB300020 and No. T1DB200010)Ministry of Railways Science and Technology Research Program ( No. 2006X023 and No. 2010X008)
文摘With the development of artificial intelligence, communication, computer and other related technologies, it becomes feasible to rebuild traditional railway with such advanced technologies in order to establish a new generation railway transport system. The railway intelligent transportation system is the trend of railway transportation system in China, and it is also the research focus of international railway transport industry. This paper presents the definition, characters, architecture, key technologies and developing pattern of the RITS(railway intelligent transportation system). Then three typical applications are introduced. Finally, the prospect of the RITS is summarized.
基金funded by ‘‘a group of four’’ Safety Science and Technology Project of State Production Safety Supervision Administration of China (No. 20130801)
文摘The research studied the influences of high temperature, high pressure, high humidity, noise and other harmful factors in mining conditions on the people health and safety, and investigated the impacts of confined environmental on human physiology factors, including temperature, humidity, noise, pressure,toxic and harmful gases in terms of environmental characteristics in underground mines and an artificial intelligence system for simulation of the environment in a confined space of deep mines. Our results show that the systolic pressure, diastolic pressure, mean pressure, heart rate, respiratory rate, typing test speed and memory level percentage are negatively correlated with temperature value, and positively correlated with humidity value; the human temperature and weight are positively correlated with temperature value, and negatively correlated with humidity value. This research lays the foundation for the study of interaction between the deep confined space environment and safety behavior.
文摘This paper will discuss the computerised developm en t control and approval system being developed for the Planning and Development C ontrol Department, City Hall of Kuala Lumpur, with stress on the GIS architectur e developed within the system. The prospects and challenges towards implementati on of the system are also discussed.
文摘The application of reinforcement learning is widely used by multi-agent systems in recent years. An agent uses a multi-agent system to cooperate with other agents to accomplish the given task, and one agent′s behavior usually affects the others′ behaviors. In traditional reinforcement learning, one agent takes the others location, so it is difficult to consider the others′ behavior, which decreases the learning efficiency. This paper proposes multi-agent reinforcement learning with cooperation based on eligibility traces, i.e. one agent estimates the other agent′s behavior with the other agent′s eligibility traces. The results of this simulation prove the validity of the proposed learning method.
