Nonlinear direct data-driven control for UAV formation flight system

This paper proposes the nonlinear direct data-driven control from theoretical analysis and practical engineering,i.e.,unmanned aerial vehicle(UAV)formation flight system.Firstly,from the theoretical point of view,consider one nonlinear closedloop system with a nonlinear plant and nonlinear feed-forward controller simultaneously.To avoid the complex identification process for that nonlinear plant,a nonlinear direct data-driven control strategy is proposed to design that nonlinear feed-forward controller only through the input-output measured data sequence directly,whose detailed explicit forms are model inverse method and approximated analysis method.Secondly,from the practical point of view,after reviewing the UAV formation flight system,nonlinear direct data-driven control is applied in designing the formation controller,so that the followers can track the leader’s desired trajectory during one small time instant only through solving one data fitting problem.Since most natural phenomena have nonlinear properties,the direct method must be the better one.Corresponding system identification and control algorithms are required to be proposed for those nonlinear systems,and the direct nonlinear controller design is the purpose of this paper.

Leader-following coordination of multiple UUVs formation under two independent topologies and time-varying delays

The new method which uses the consensus algorithm to solve the coordinate control problems of multiple unmanned underwater vehicles （multi-UUVs） formation in the case of leader-following is adapted. As the communication between the UUVs is difficult and it is easy to be interfered under the water, time delay is assumed to be time-varying during the members communicate with each other. Meanwhile, the state feedback linearization method is used to transfer the nonlinear and coupling model of UUV into double-integrator dynamic. With this simplified double-integrator math model, the UUV formation coordinate control is regarded as consensus problem with time-varying communication delays. In addition, the position and velocity topologies are adapted to reduce the data volume in each data packet which is sent between members in formation. With two independent topologies designed, two cases of communication delay which are same and different are considered and the sufficient conditions are proposed and analyzed. The stability of the multi-UUVs formation is proven by using Lyapunov-Razumilkhin theorem. Finally, the simulation results are presented to confirm and illustrate the theoretical results.

Long-term effects of main-body's obliquity on satellite formation perturbed by third-body gravity in elliptical and inclined orbit

A new non-simplified model of formation flying is derived in the presence of an oblate main- body and third-body perturbation. In the proposed model, considering the perturbation of the third- body in an inclined orbit, the effect of obliquity （axial tilt） of the main-body is becoming important and has been propounded in the absolute motion of a reference satellite and the relative motion of a follower satellite. From a new point of view, J2 perturbed relative motion equations and considering a disturbing body in an elliptic inclined three dimensional orbit, are derived using Lagrangian mechanics based on accurate introduced perturbed reference satellite motion. To validate the accuracy of the model presented in this study, an auxiliary model was constructed as the Main-body Center based Relative Motion （MCRM） model. Finally, the importance of the main-body＇s obliquity is demonstrated by several examples related to the Earth-Moon system in relative motion and lunar satellite formation keeping. The main-body＇s obliquity has a remarkable effect on formation keeping in the examined in-track and projected circular orbit （PCO） formations.

Cooperative decision-making algorithm with efficient convergence for UCAV formation in beyond-visual-range air combat based on multi-agent reinforcement learning

Highly intelligent Unmanned Combat Aerial Vehicle(UCAV)formation is expected to bring out strengths in Beyond-Visual-Range(BVR)air combat.Although Multi-Agent Reinforcement Learning(MARL)shows outstanding performance in cooperative decision-making,it is challenging for existing MARL algorithms to quickly converge to an optimal strategy for UCAV formation in BVR air combat where confrontation is complicated and reward is extremely sparse and delayed.Aiming to solve this problem,this paper proposes an Advantage Highlight Multi-Agent Proximal Policy Optimization(AHMAPPO)algorithm.First,at every step,the AHMAPPO records the degree to which the best formation exceeds the average of formations in parallel environments and carries out additional advantage sampling according to it.Then,the sampling result is introduced into the updating process of the actor network to improve its optimization efficiency.Finally,the simulation results reveal that compared with some state-of-the-art MARL algorithms,the AHMAPPO can obtain a more excellent strategy utilizing fewer sample episodes in the UCAV formation BVR air combat simulation environment built in this paper,which can reflect the critical features of BVR air combat.The AHMAPPO can significantly increase the convergence efficiency of the strategy for UCAV formation in BVR air combat,with a maximum increase of 81.5%relative to other algorithms.

Growth of nitrate contribution to aerosol pollution during wintertime in Xi'an,northwest China:Formation mechanism and effects of NH_(3)

With the strengthened controls on SO2 emissions and extensive increases in motor vehicles'exhaust,aerosol pollution shifts from sulfate-rich to nitrate-rich in recent years in Xi'an,China.To further gain insights into the factors on nitrate formation and efficiently mitigate air pollution,highly time-resolved observations of water-soluble inorganic ions(WSIIs)in PM_(2.5) were measured in a suburban area of Xi'an,China during wintertime.Hourly concentration of total WSIIs is 39.8μg m-3 on average,accounting for 50.3%of PM_(2.5) mass.In contrast to a slight decrease in the mass fraction of SO_(4)^(2-),NO_(3)-shows a sig-nificant increase of the PM_(2.5) contribution with the aggravation of aerosol pollution.This suggests the importance of NO_(3)-formation to haze evolution.Furthermore,homogeneous reactions govern the formation of NO_(3)-,while alkali metals such as calcium and sodium play an additional role in retaining NO_(3)-in PM_(2.5) during clean periods.However,the heterogeneous hydrolysis reaction contributed more to NO_(3)-formation during the pollution periods under high relative humidity.Our investigation reveals that temperature,relative humidity,oxidant,and ammonia emissions facilitate rapid NO_(3)-formation.Using the random forest(RF)model,NO_(3)-concentrations were successfully simulated with measured variables for the training and testing datasets(R2>0.95).Among these variables,CO,NH_(3),and NO_(2) were found to be the main factors affecting the NO_(3)-concentrations.Compared with the period without vehicle re-striction,the contributions of NO_(3)-and NH4+to PM_(2.5) mass decreased by 5.3%and 3.4%in traffic re-striction periods,respectively.The vehicle restriction leads to the decreases of precursor gases of NO_(2),SO_(2),and NH_(3) by 12.8%,5.9%,and 27.6%,respectively.The results demonstrate collaborative emission reduction of NO_(x) and NH_(3) by vehicle restrictions,and using new energy vehicles(or electric vehicles)can effectively alleviate particulate matter pollution in northwest China.

Carbonyl emissions from heavy-duty diesel vehicle exhaust in China and the contribution to ozone formation potential

Fifteen heavy-duty diesel vehicles were tested on chassis dynamometer by using typical heavy duty driving cycle and fuel economy cycle. The air from the exhaust was sampled by 2,4- dinitrophenyhydrazine cartridge and 23 carbonyl compounds were analyzed by high performance liquid chromatography. The average emission factor of carbonyls was 97.2 mg/km, higher than that of light-duty diesel vehicles and gasoline-powered vehicles. Formaldehyde, acetaldehyde, acetone and propionaidehyde were the species with the highest emission factors. Main influencing factors for carbonyl emissions were vehicle type, average speed and regulated emission standard, and the impact of vehicle loading was not evident in this study. National emission of carbonyls from diesel vehicles exhaust was calculated for China, 2011, based on both vehicle miles traveled and fuel consumption. Carbonyl emission of diesel vehicle was estimated to be 45.8 Gg, and was comparable to gasolinepowered vehicles （58.4 Gg）. The emissions of formaldehyde, acetaldehyde and acetone were 12.6, 6.9, 3.8 Gg, respectively. The ozone formation potential of carbonyls from diesel vehicles exhaust was 537 mg O3/km, higher than 497 mg O3/km of none-methane hydrocarbons emitted from diesel vehicles.

On-board measurements of gaseous pollutant emission characteristics under real driving conditions from light-duty diesel vehicles in Chinese cities

A total of 15 light-duty diesel vehicles（LDDVs） were tested with the goal of understanding the emission factors of real-world vehicles by conducting on-board emission measurements. The emission characteristics of hydrocarbons（HC） and nitrogen oxides（NOx） at different speeds, chemical species profiles and ozone formation potential（OFP） of volatile organic compounds（VOCs） emitted from diesel vehicles with different emission standards were analyzed. The results demonstrated that emission reductions of HC and NOxhad been achieved as the control technology became more rigorous from Stage I to Stage IV. It was also found that the HC and NOxemissions and percentage of O2 dropped with the increase of speed, while the percentage of CO2 increased. The abundance of alkanes was significantly higher in diesel vehicle emissions, approximately accounting for 41.1%–45.2%, followed by aromatics and alkenes. The most abundant species were propene,ethane, n-decane, n-undecane, and n-dodecane. The maximum incremental reactivity（MIR）method was adopted to evaluate the contributions of individual VOCs to OFP. The results indicated that the largest contributors to O3 production were alkenes and aromatics, which accounted for 87.7%–91.5%. Propene, ethene, 1,2,4-trimethylbenzene, 1-butene, and1,2,3-trimethylbenzene were the top five VOC species based on their OFP, and accounted for 54.0%-64.8% of the total OFP. The threshold dilution factor was applied to analyze the possibility of VOC stench pollution. The majority of stench components emitted from vehicle exhaust were aromatics, especially p-diethylbenzene, propylbenzene, m-ethyltoluene, and p-ethyltoluene.