Recent advancements in reinforcement learning(RL)and computational resources have demonstrated the efficacy of data-driven methodologies for robotic locomotion control and physical design opti-mization,providing a sca...Recent advancements in reinforcement learning(RL)and computational resources have demonstrated the efficacy of data-driven methodologies for robotic locomotion control and physical design opti-mization,providing a scalable alternative to traditional human-crafted design paradigms.However,existing co-design approaches face a critical challenge:the computational intractability of exploring high-dimensional design spaces,exacerbated by the resource-intensive nature of policy training and candidate design evaluations.To address this limitation,we propose an efficient co-adaptation frame-work for humanoid robot kinematics optimization.Building on a bi-level optimization architecture that jointly optimizes mechanical designs and control policies,our method achieves computational efficiency through two synergistic strategies:(1)a universal policy generalizable across design variations,and(2)a surrogate-assisted fitness evaluation mechanism.We implement the method with humanoid robot Kuafu,and by experimental results we demonstrate the proposed method effectively reduces the cost and the optimized design can achieve near-optimal performance.展开更多
In order to seek the co-adaptability solution to conflict events in construction engineering projects, a new method referred to as segmented hierarchical algorithm is proposed in this paper by means of comparing co-ad...In order to seek the co-adaptability solution to conflict events in construction engineering projects, a new method referred to as segmented hierarchical algorithm is proposed in this paper by means of comparing co-adaptability evolution process of conflict events to the stackelberg model. By this new algorithm, local solutions to the first-order transformation of co-adaptability for conflict events can be obtained, based upon which, a global solution to the second-order transformation of co-adaptability for conflict events can also be decided by judging satisfaction degree of local solutions. The research results show that this algorithm can be used not only for obtaining co-adaptability solution to conflict events efficiently, but also for other general decision-making problems with multi-layers and multi-subsidi-aries in project management field.展开更多
The precision and the extent of behavioral thermoregulation are likely to provide fitness benefits to ectotherms.Yet the factors driving variation in selected or preferred body temperature(T_(set))and its usefulness a...The precision and the extent of behavioral thermoregulation are likely to provide fitness benefits to ectotherms.Yet the factors driving variation in selected or preferred body temperature(T_(set))and its usefulness as a proxy for optimal physiological temperature(Topt)are still debated.Although T_(set)is often conserved among closely related species,substantial variation at the individual,population and species level has also been reported.However,the repeatability(calculated as the intra-class correlation coefficient)of T_(set)is generally low.One factor that influences T_(set)is feeding status,with fed reptiles typically showing higher T_(set),a process thought to aid meal digestion.Here,using experiments simulating realistic feeding and fasting regimes in Agama atra,a heliothermic lizard from southern Africa,we test if T_(set)and its repeatability under these 2 states significantly differ.Daily T_(set)ranged from 33.7 to 38.4℃,with a mean(±SE)of 36.7±0.1℃ for fed and 36.6±0.1℃ for unfed individuals.Comparisons of repeatability showed that females tend to be more consistent in the selection of body temperature than males,but not significantly so regardless of feeding status.We report some of the highest repeatability estimates of T_(set)to date(full range:0.229–0.642),and that the weak positive effects of feeding status on T_(set)did not increase its repeatability.In conclusion,one of the major prerequisites for natural selection,consistent among-individual variation,is present,making the adaptive significance of T_(set)considerably more plausible.展开更多
基金supported in part by the National Natural Science Foundation of China(62403060)in part by the Postdoctoral Fellowship Program of CPSF(GZC20233399)+1 种基金in part by the Beijing Natural Science Foundation(L243004)in part by the"111"Project(B08043).
文摘Recent advancements in reinforcement learning(RL)and computational resources have demonstrated the efficacy of data-driven methodologies for robotic locomotion control and physical design opti-mization,providing a scalable alternative to traditional human-crafted design paradigms.However,existing co-design approaches face a critical challenge:the computational intractability of exploring high-dimensional design spaces,exacerbated by the resource-intensive nature of policy training and candidate design evaluations.To address this limitation,we propose an efficient co-adaptation frame-work for humanoid robot kinematics optimization.Building on a bi-level optimization architecture that jointly optimizes mechanical designs and control policies,our method achieves computational efficiency through two synergistic strategies:(1)a universal policy generalizable across design variations,and(2)a surrogate-assisted fitness evaluation mechanism.We implement the method with humanoid robot Kuafu,and by experimental results we demonstrate the proposed method effectively reduces the cost and the optimized design can achieve near-optimal performance.
基金Soft Science Foundation of Shaanxi Province of China (Grant No. 2005KR121)Natural Science Foundation of Shaanxi Prov-ince of China (Grant No. 2005E214)
文摘In order to seek the co-adaptability solution to conflict events in construction engineering projects, a new method referred to as segmented hierarchical algorithm is proposed in this paper by means of comparing co-adaptability evolution process of conflict events to the stackelberg model. By this new algorithm, local solutions to the first-order transformation of co-adaptability for conflict events can be obtained, based upon which, a global solution to the second-order transformation of co-adaptability for conflict events can also be decided by judging satisfaction degree of local solutions. The research results show that this algorithm can be used not only for obtaining co-adaptability solution to conflict events efficiently, but also for other general decision-making problems with multi-layers and multi-subsidi-aries in project management field.
基金The research was conducted with the approval of the Animal Ethics Committee of Stellenbosch University(SU-ACUD14-00110)the Cape Nature permit 0056-AAA007-00206Funding was provided by an NRF Competitive Programme for Rated Researchers’grant to SCT.
文摘The precision and the extent of behavioral thermoregulation are likely to provide fitness benefits to ectotherms.Yet the factors driving variation in selected or preferred body temperature(T_(set))and its usefulness as a proxy for optimal physiological temperature(Topt)are still debated.Although T_(set)is often conserved among closely related species,substantial variation at the individual,population and species level has also been reported.However,the repeatability(calculated as the intra-class correlation coefficient)of T_(set)is generally low.One factor that influences T_(set)is feeding status,with fed reptiles typically showing higher T_(set),a process thought to aid meal digestion.Here,using experiments simulating realistic feeding and fasting regimes in Agama atra,a heliothermic lizard from southern Africa,we test if T_(set)and its repeatability under these 2 states significantly differ.Daily T_(set)ranged from 33.7 to 38.4℃,with a mean(±SE)of 36.7±0.1℃ for fed and 36.6±0.1℃ for unfed individuals.Comparisons of repeatability showed that females tend to be more consistent in the selection of body temperature than males,but not significantly so regardless of feeding status.We report some of the highest repeatability estimates of T_(set)to date(full range:0.229–0.642),and that the weak positive effects of feeding status on T_(set)did not increase its repeatability.In conclusion,one of the major prerequisites for natural selection,consistent among-individual variation,is present,making the adaptive significance of T_(set)considerably more plausible.
