The objective of this work is to develop an innovative system(ROSGPT)that merges large language models(LLMs)with the robot operating system(ROS),facilitating natural language voice control of mobile robots.This integr...The objective of this work is to develop an innovative system(ROSGPT)that merges large language models(LLMs)with the robot operating system(ROS),facilitating natural language voice control of mobile robots.This integration aims to bridge the gap between human-robot interaction(HRI)and artificial intelligence(AI).ROSGPT integrates several subsystems,including speech recognition,prompt engineering,LLM and ROS,enabling seamless control of robots through human voice or text commands.The LLM component is optimized,with its performance refined from the open-source Llama2 model through fine-tuning and quantization procedures.Through extensive experiments conducted in both real-world and virtual environments,ROSGPT demonstrates its efficacy in meeting user requirements and delivering user-friendly interactive experiences.The system demonstrates versatility and adaptability through its ability to comprehend diverse user commands and execute corresponding tasks with precision and reliability,thereby showcasing its potential for various practical applications in robotics and AI.The demonstration video can be viewed at https://iklxo6z9yv.feishu.cn/docx/Lux3dmTDxoZ5YnxWJTZcxUCWnTh.展开更多
Assessment of ecological flow or water level for water bodies is important for the protection of de- graded or degrading ecosystems caused by water shortage in arid regions, and it has become a key issue in water reso...Assessment of ecological flow or water level for water bodies is important for the protection of de- graded or degrading ecosystems caused by water shortage in arid regions, and it has become a key issue in water resources planning. In the past several decades, many methods have been proposed to assess ecological flow for rivers and ecological water level for lakes or wetlands. To balance water uses by human and ecosystems, we proposed a general multi-objective programming model to determine minimum ecological flow or water level for inland water bodies, where two objectives are water index for human and habitat index for ecosystems, respectively Using the weighted sum method for multi-objective optimization, minimum ecological flow or water level can be determined from the breakpoint in the water index-habitat index curve, which is similar to the slope method to de- termine minimum ecological flow from wetted perimeter-discharge curve. However, the general multi-objective programming model is superior to the slope method in its physical meaning and calculation method. This model provides a general analysis method for ecological water uses of different inland water bodies, and can be used to define minimum ecological flow or water level by choosing appropriate water and habitat indices. Several com- monly used flow or water level assessment methods were found to be special cases of the general model, including the wetted perimeter method and the multi-objective physical habitat simulation method for ecological river flow, the inundated forest width method for regeneration flow of floodplain forest and the lake surface area method for eco- logical lake level. These methods were applied to determine minimum ecological flow or water level for two repre- sentative rivers and a lake in northern Xinjiang of China, including minimum ecological flow for the Ertix River, minimum regeneration flow for floodplain forest along the midstream of Kaxgar River, and minimum ecological lake level for the Ebinur Lake. The results illustrated the versatility of the general model, and can provide references for water resources planning and ecosystem protection for these rivers and lake.展开更多
基金National Natural Science Foundation of China(No.61601112)。
文摘The objective of this work is to develop an innovative system(ROSGPT)that merges large language models(LLMs)with the robot operating system(ROS),facilitating natural language voice control of mobile robots.This integration aims to bridge the gap between human-robot interaction(HRI)and artificial intelligence(AI).ROSGPT integrates several subsystems,including speech recognition,prompt engineering,LLM and ROS,enabling seamless control of robots through human voice or text commands.The LLM component is optimized,with its performance refined from the open-source Llama2 model through fine-tuning and quantization procedures.Through extensive experiments conducted in both real-world and virtual environments,ROSGPT demonstrates its efficacy in meeting user requirements and delivering user-friendly interactive experiences.The system demonstrates versatility and adaptability through its ability to comprehend diverse user commands and execute corresponding tasks with precision and reliability,thereby showcasing its potential for various practical applications in robotics and AI.The demonstration video can be viewed at https://iklxo6z9yv.feishu.cn/docx/Lux3dmTDxoZ5YnxWJTZcxUCWnTh.
基金supported by the Open Research Fund Program of State key Laboratory of Hydroscience and Engineering, Tsinghua University (sklhse-2013-A-03)the National Natural Science Foundation of China (50879041)
文摘Assessment of ecological flow or water level for water bodies is important for the protection of de- graded or degrading ecosystems caused by water shortage in arid regions, and it has become a key issue in water resources planning. In the past several decades, many methods have been proposed to assess ecological flow for rivers and ecological water level for lakes or wetlands. To balance water uses by human and ecosystems, we proposed a general multi-objective programming model to determine minimum ecological flow or water level for inland water bodies, where two objectives are water index for human and habitat index for ecosystems, respectively Using the weighted sum method for multi-objective optimization, minimum ecological flow or water level can be determined from the breakpoint in the water index-habitat index curve, which is similar to the slope method to de- termine minimum ecological flow from wetted perimeter-discharge curve. However, the general multi-objective programming model is superior to the slope method in its physical meaning and calculation method. This model provides a general analysis method for ecological water uses of different inland water bodies, and can be used to define minimum ecological flow or water level by choosing appropriate water and habitat indices. Several com- monly used flow or water level assessment methods were found to be special cases of the general model, including the wetted perimeter method and the multi-objective physical habitat simulation method for ecological river flow, the inundated forest width method for regeneration flow of floodplain forest and the lake surface area method for eco- logical lake level. These methods were applied to determine minimum ecological flow or water level for two repre- sentative rivers and a lake in northern Xinjiang of China, including minimum ecological flow for the Ertix River, minimum regeneration flow for floodplain forest along the midstream of Kaxgar River, and minimum ecological lake level for the Ebinur Lake. The results illustrated the versatility of the general model, and can provide references for water resources planning and ecosystem protection for these rivers and lake.
