Cooperative multi-target hunting by unmanned surface vehicles based on multi-agent reinforcement learning

To solve the problem of multi-target hunting by an unmanned surface vehicle(USV)fleet,a hunting algorithm based on multi-agent reinforcement learning is proposed.Firstly,the hunting environment and kinematic model without boundary constraints are built,and the criteria for successful target capture are given.Then,the cooperative hunting problem of a USV fleet is modeled as a decentralized partially observable Markov decision process(Dec-POMDP),and a distributed partially observable multitarget hunting Proximal Policy Optimization(DPOMH-PPO)algorithm applicable to USVs is proposed.In addition,an observation model,a reward function and the action space applicable to multi-target hunting tasks are designed.To deal with the dynamic change of observational feature dimension input by partially observable systems,a feature embedding block is proposed.By combining the two feature compression methods of column-wise max pooling(CMP)and column-wise average-pooling(CAP),observational feature encoding is established.Finally,the centralized training and decentralized execution framework is adopted to complete the training of hunting strategy.Each USV in the fleet shares the same policy and perform actions independently.Simulation experiments have verified the effectiveness of the DPOMH-PPO algorithm in the test scenarios with different numbers of USVs.Moreover,the advantages of the proposed model are comprehensively analyzed from the aspects of algorithm performance,migration effect in task scenarios and self-organization capability after being damaged,the potential deployment and application of DPOMH-PPO in the real environment is verified.

Genomewide association study of Aegilops tauschii traits under seedling-stage cadmium stress

Aegilops tauschii Ais a wild relative of common wheat(Triticum aestivum) and acts as an important resource of elite genes including genes for resistance to biotic and abiotic stresses. To improve the cadmium(Cd) tolerance of wheat varieties using A. tauschii resources, we investigated the genetic variation of biomass-based Cd tolerance in 235 A. tauschii accessions treated with 0(control) and100 μmol L-1CdC l2(as Cd stress). Simultaneously, we performed a genomewide association study(GWAS) using a single-nucleotide polymorphism chip containing 7185 markers. Six markers were found to be significantly associated with Cd tolerance by a general linear model and a mixed linear model. These markers were close to several candidate/flanking genes associated with Cd tolerance according to results in public databases, including pdil5-1, Acc-1, DME-5A,TaA P2-D, TaA P2-B, Vrn-B1, and FtsH-like protein gene. The A. tauschii accessions were classified as high, moderate, and low Cd-tolerant according to a secondary index, the synthetic index(SI), in proportions of 9%, 57%, and 34%, respectively. By the average SI, accessions from Afghanistan,Turkey, Azerbaijan, and Iran showed relatively high Cd tolerance.

Cellular Automata Simulations of Random Pitting Process on Steel Reinforcement Surface

The corrosion of reinforcement in the concrete will cause the effective cross-sectional area of reinforcement to be weakened and the performance of reinforcement to change and lead to the degradation of the bond behavior between reinforcement and concrete,which can seriously affect the mechanical properties of the structural elements.Therefore,it is of great practical significance to accurately simulate the corrosion morphology and the corrosion products of reinforcement.This paper improves the previous cellular automata models and establishes a new cellular automata model framework for simulating the random pitting corrosion process of reinforcement in concrete.This model defines the detailed local evolution laws of material transformation,penetration and diffusion processes during the corrosion.Meanwhile,based on the spatial inhomogeneity of corrosion,three parameters are introduced into the model:The dissolution probability parameter p,the local corrosion space parameterλand the local corrosion probability parameterε,which establishes a parameterized model of corrosion probability.The research results show that the common steel reinforcement corrosion morphology can be obtained by adjusting the parameters.The volume expansion rate of the corrosion products is about 2,which is consistent with the relevant experimental research results.The cellular automata model in this paper can simulate the common steel reinforcement corrosion morphology and corrosion products in engineering.

Highly selectiveα-C(sp^(3))-H arylation of alkenyl amides via nickel chain-walking catalysis

Herein,we report the migratory hydroarylation of unactivated alkenes with aryl iodides using native and weakly coordinating amide directors under mild conditions.Synergistic coordination of the monodentate directing group and the ligand enable the highly regioselective migratory hydroarylation via a chain walking process to form the thermodynamically stable five-membered nickelacyle intermediate.The protocol provides a variety of valuableα-aryl-substituted alkylamine products,and exhibited good functional group tolerance.The modification of bioactive compounds such as fenofibrate and indomethacin further highlights the synthetic value of this protocol.