Autonomous vehicles with self-evolution capabilities are expected to improve their performance through learning algorithms,to automatically adapt to the external environment.However,due to the infinity,complexity,and ...Autonomous vehicles with self-evolution capabilities are expected to improve their performance through learning algorithms,to automatically adapt to the external environment.However,due to the infinity,complexity,and variability of the actual traffic environment,it is necessary to develop quantitative representation indicators of scenario difficulty and generate targeted scenarios to ensure the evolution gradually,so as to quickly approach the performance limit of the algorithm.Therefore,this paper proposes a data-driven quantitative representation method of scenario difficulty.Specifically,the concept of environment agent is proposed,and a reinforcement learning method combined with mechanism knowledge is constructed for policy search to obtain an agent with an adversarial behavior.The model parameters of the environment agent at different stages in the training process are extracted to construct a policy group,and then agents with different adversarial intensities are obtained,which are used to realize data generation in different difficulty scenarios through the simulation environment.Finally,a data-driven scenario difficulty quantitative representation model is constructed,which is used to output the environment agent policy under different difficulties.Experimental results show the effectiveness of the proposed method.The result analysis shows that the proposed algorithm can generate reasonable and interpretable scenarios with high discrimination and can provide quantifiable difficulty representation without any expert logic rule design.Compared with the rule-based discrete scenario difficulty representation method,the proposed algorithm can achieve continuous difficulty representation.The video link is https://www.youtube.com/watch?v=GceGdqAm9Ys.展开更多
China,which has already introduced an environmental tax in an effort to decarbonize,has recently begun emissions trading and is using two environmental policies in tandem,but there are concerns about the impact on gro...China,which has already introduced an environmental tax in an effort to decarbonize,has recently begun emissions trading and is using two environmental policies in tandem,but there are concerns about the impact on growth and trade.Trade and environmental policies affect firms'entry and exit,resulting in changes in aggregate productivity and pollution emissions.This study compares the impacts of single regulation and dual regulation on welfare,using a research-and-development based growth model with heterogeneous firms.Under single regulation,the cleansing effect of trade liberalization could be undermined.Under dual regulation,trade liberalization decreases pollution and improves average productivity whereas decreasing total permits reduces pollution.From the perspective of improving welfare it is desirable to choose dual regulation because trade liberalization can reduce total pollution emissions via the cleansing effect of trade liberalization.展开更多
基金the National Key R&D Program of China under Grant No.2022YFB2502900the National Natural Science Foundation of China(Grant Number:U23B2061)+1 种基金the Fundamental Research Funds for the Central Universities of Chinathe Xiaomi Young Talent Program,and we thank the reviewers for the valuable suggestions.
文摘Autonomous vehicles with self-evolution capabilities are expected to improve their performance through learning algorithms,to automatically adapt to the external environment.However,due to the infinity,complexity,and variability of the actual traffic environment,it is necessary to develop quantitative representation indicators of scenario difficulty and generate targeted scenarios to ensure the evolution gradually,so as to quickly approach the performance limit of the algorithm.Therefore,this paper proposes a data-driven quantitative representation method of scenario difficulty.Specifically,the concept of environment agent is proposed,and a reinforcement learning method combined with mechanism knowledge is constructed for policy search to obtain an agent with an adversarial behavior.The model parameters of the environment agent at different stages in the training process are extracted to construct a policy group,and then agents with different adversarial intensities are obtained,which are used to realize data generation in different difficulty scenarios through the simulation environment.Finally,a data-driven scenario difficulty quantitative representation model is constructed,which is used to output the environment agent policy under different difficulties.Experimental results show the effectiveness of the proposed method.The result analysis shows that the proposed algorithm can generate reasonable and interpretable scenarios with high discrimination and can provide quantifiable difficulty representation without any expert logic rule design.Compared with the rule-based discrete scenario difficulty representation method,the proposed algorithm can achieve continuous difficulty representation.The video link is https://www.youtube.com/watch?v=GceGdqAm9Ys.
基金the Early Career Scientists from the Japan Society for the Promotion of Science Nos.19K13706 and 22K13409(JSPS KAKENHI Grant Nos.JP19K13706 and JP22K13409).
文摘China,which has already introduced an environmental tax in an effort to decarbonize,has recently begun emissions trading and is using two environmental policies in tandem,but there are concerns about the impact on growth and trade.Trade and environmental policies affect firms'entry and exit,resulting in changes in aggregate productivity and pollution emissions.This study compares the impacts of single regulation and dual regulation on welfare,using a research-and-development based growth model with heterogeneous firms.Under single regulation,the cleansing effect of trade liberalization could be undermined.Under dual regulation,trade liberalization decreases pollution and improves average productivity whereas decreasing total permits reduces pollution.From the perspective of improving welfare it is desirable to choose dual regulation because trade liberalization can reduce total pollution emissions via the cleansing effect of trade liberalization.
