Oilseed rape is one of the most important oil crops globally.Attaining the appropriate cultivation method(planting pattern and nitrogen level)is necessary to achieve high yield,quality and resource utilization efficie...Oilseed rape is one of the most important oil crops globally.Attaining the appropriate cultivation method(planting pattern and nitrogen level)is necessary to achieve high yield,quality and resource utilization efficiency.However,the optimal method for oilseed rape varies across countries and regions.The objective of the present study was to determine an appropriate cultivation method,including planting pattern and nitrogen application,for winter oilseed rape in northwestern China.Two planting patterns:ridge film mulching and furrow planting(RFMF)and flat planting(FP),and six nitrogen(N)amounts:0(N0),60(N60),120(N120),180(N180),240(N240),and 300(N300)kg N ha–1 were applied across three growing seasons(2014–2017).Three comprehensive decision analysis methods:principal component analysis,grey correlation degree analysis and the combined entropy weight and dynamic technique for order preference by similarity to ideal solution method were used to evaluate the growth and physiological indicators,nutrient uptake,yield,quality,evapotranspiration,and water use efficiency of winter oilseed rape.Planting pattern,nitrogen amount and their interaction significantly affected the indicators aforementioned.The RFMF pattern significantly increased all indicators over the FP pattern.Application of N also markedly increased all the indicators except for seed oil content,but the yield,oil production and water use efficiency were decreased when N fertilizer exceeded 180 kg N ha–1 under FP and 240 kg N ha–1 under RFFM.The evaluation results of the three comprehensive decision analysis methods indicated that RFMF planting pattern with 240 kg N ha–1 is an appropriate cultivation method for winter oilseed rape in northwestern China.These findings are of vital significance to maximize yield,optimize quality and improve resource use efficiencies of winter oilseed rape.展开更多
After detecting a target object,a service robot must approach the target object to perform the associated service task.In active object detection(AOD)tasks,effective feature information representation and comprehensiv...After detecting a target object,a service robot must approach the target object to perform the associated service task.In active object detection(AOD)tasks,effective feature information representation and comprehensive action execution strategies are crucial.Currently,most AOD tasks are accomplished by traditional reinforcement learning algorithms,but there are still problems such as high task failure rates and model training efficiency.To solve these problems,this paper proposes a combined data-driven and knowledge-guided solution.First,semantic information features,depth information features and target object bounding box information are used as inputs to comprehensively represent feature information.Second,a policy network is constructed based on the proximal policy optimizaton(PPO)algorithm.The reward value is set according to the robot′s action,the position of the bounding box,and the distance to the target object,and then applied to the robot′s training process.Finally,the knowledge of the path experience in the task,the robot′s collision avoidance ability and the prediction of target object loss are combined to guide the robot′s behavior,and a comprehensive decision model is proposed to enable the robot to make the best decision.Relevant experiments were conducted on an active vision dataset.The robot achieves an average success rate of 91.36%and an average step size of 9.3631 in performing the AOD task in the test scenes,which verifies the effectiveness of the proposed scheme.展开更多
基金This research was supported by the Special Fund forAgroscientific Research in the Public Interest,China(201503125 and 201503105)the Fundamental Research Funds for the Central Universities,China(2452018089).
文摘Oilseed rape is one of the most important oil crops globally.Attaining the appropriate cultivation method(planting pattern and nitrogen level)is necessary to achieve high yield,quality and resource utilization efficiency.However,the optimal method for oilseed rape varies across countries and regions.The objective of the present study was to determine an appropriate cultivation method,including planting pattern and nitrogen application,for winter oilseed rape in northwestern China.Two planting patterns:ridge film mulching and furrow planting(RFMF)and flat planting(FP),and six nitrogen(N)amounts:0(N0),60(N60),120(N120),180(N180),240(N240),and 300(N300)kg N ha–1 were applied across three growing seasons(2014–2017).Three comprehensive decision analysis methods:principal component analysis,grey correlation degree analysis and the combined entropy weight and dynamic technique for order preference by similarity to ideal solution method were used to evaluate the growth and physiological indicators,nutrient uptake,yield,quality,evapotranspiration,and water use efficiency of winter oilseed rape.Planting pattern,nitrogen amount and their interaction significantly affected the indicators aforementioned.The RFMF pattern significantly increased all indicators over the FP pattern.Application of N also markedly increased all the indicators except for seed oil content,but the yield,oil production and water use efficiency were decreased when N fertilizer exceeded 180 kg N ha–1 under FP and 240 kg N ha–1 under RFFM.The evaluation results of the three comprehensive decision analysis methods indicated that RFMF planting pattern with 240 kg N ha–1 is an appropriate cultivation method for winter oilseed rape in northwestern China.These findings are of vital significance to maximize yield,optimize quality and improve resource use efficiencies of winter oilseed rape.
基金supported in part by the National Natural Science Foundation of China(Nos.62273203 and U1813215)in part by the Special Fund for the Taishan Scholars Program of Shandong Province,China(No.ts2015110005).
文摘After detecting a target object,a service robot must approach the target object to perform the associated service task.In active object detection(AOD)tasks,effective feature information representation and comprehensive action execution strategies are crucial.Currently,most AOD tasks are accomplished by traditional reinforcement learning algorithms,but there are still problems such as high task failure rates and model training efficiency.To solve these problems,this paper proposes a combined data-driven and knowledge-guided solution.First,semantic information features,depth information features and target object bounding box information are used as inputs to comprehensively represent feature information.Second,a policy network is constructed based on the proximal policy optimizaton(PPO)algorithm.The reward value is set according to the robot′s action,the position of the bounding box,and the distance to the target object,and then applied to the robot′s training process.Finally,the knowledge of the path experience in the task,the robot′s collision avoidance ability and the prediction of target object loss are combined to guide the robot′s behavior,and a comprehensive decision model is proposed to enable the robot to make the best decision.Relevant experiments were conducted on an active vision dataset.The robot achieves an average success rate of 91.36%and an average step size of 9.3631 in performing the AOD task in the test scenes,which verifies the effectiveness of the proposed scheme.
