The Top10 engineering research frontiers in the field of mechanical and delivery engineering (hereinafter referred to as the mechanical field) involve mechanical engineering, ship and Marine engineering, aerospace science and technology, weapons science and technology, power and electrical equipment engineering and technology, transportation engineering and other disciplines. Among them, the "adaptive tracking of autonomous underwater vehicle" "consistency of multi-agent system control" "adaptive neural network control of the manipulator" "underwater autonomous navigation system", "global navigation satellite system optimization" and "offshore wave energy resources assessment and use" is the development of the traditional research, "lithium ion battery thermal management technology" "cognitive wireless network" "for target recognition based on the sense of touch" and "electrical/magnetic strengthen nano fluid convection heat transfer" is a new frontier. The paper has been published year by year from 2012 to 2017. "lithium ion battery thermal management technology" and "electric/magnetic field enhanced nanofluid convection heat transfer" are the most significant directions for the growth of paper publication in recent years.

Top10 engineering research frontiers in mechanical field

(1) adaptive tracking of autonomous underwater vehicles

As a typical strongly coupled nonlinear system, autonomous underwater vehicle (auv) is easily disturbed by time-varying factors such as underwater currents, and has stronger model and parameter uncertainty than general rigid body. For nonlinear systems with linearized unknown parameters, adaptive techniques are generally used to estimate unknown parameters online. For systems with non-linearized parameters, the neural network method can be used to compensate the system uncertainty and ensure the stability of tracking error. According to the different configuration of propeller, the autonomous underwater vehicle (auv) tracking research is mainly divided into under-actuated auv tracking control and full-actuated auv tracking control. The tracking error of the underactuated underwater vehicle tracking control system has a strong nonlinear coupling between the motion degrees of freedom. Since each degree of freedom of the fully actuated underwater vehicle (auv) tracking control system has an independent control input, scholars at home and abroad mainly use adaptive backstepping control method to obtain the trajectory tracking controller with global linear stability. The cooperative tracking and detection efficiency of autonomous underwater vehicle formation is much better than that of monomer tracking and detection. On the basis of gradually improving the adaptive monomer tracking and control technology of autonomous underwater vehicle, the research on adaptive formation cooperative control technology and intelligent path planning technology will be the development trend of this research direction.

(2) consistency control of multi-agent system

From human beings to birds, fish, insects, bacteria and cells, large-scale group movement exists widely in nature. Systems of interconnected and constantly moving individuals emerge in a rich variety of highly coordinated swarm dynamics. Multi-agent system is a way to understand biological and natural clustering behavior, and has considerable application value in industrial multi-robot group coordination, uav formation control, human group behavior regulation and guidance, wireless sensor network optimization and other fields. A multi-agent system is composed of a series of interacting agents. Each agent expresses the structure, function and behavior characteristics of the system through communication, cooperation, coordination, scheduling, management and control, and completes a large number of complex tasks that a single agent cannot complete. The multi-agent system has the characteristics of autonomy, distribution and coordination, as well as the abilities of self-organization, learning and reasoning. Therefore, the multi-agent system has strong robustness and reliability in solving practical problems. Because of the development of biology, computer science, artificial intelligence, automation science, physical science and so on, multi-agent system has become the frontier problem of engineering control. Inspired by the clustering phenomenon that exists widely in nature, the early research on the coordination of multi-agent system makes use of the methods of mathematics, computer simulation and system science to explore. In recent decades, there have been a lot of researches on the cooperative control theory of multi-agent systems. The basic problems of multi-agent coordination control include consistency control, agglomeration control, swarm control and formation control. Consistency control is one of the most basic problems in coordination control of multi-agent systems. Consistency control is to design a consistency protocol to make all agents interact with each other through local information and achieve the consistency of the target state value of all agents. Consistency control is mainly studied from three aspects: agent dynamics complexity, communication topology complexity and network information transmission complexity. At present, the most important application of multi-agent system is the collaboration of group robots. In particular, the traditional multi-robot production line often adopts the centralized control structure, which is difficult to adapt to the task-oriented small batch and multi-variety production, and lacks the ability of agile manufacturing. As the international manufacturing industry is changing to a large, complex, dynamic and open direction, the complex operation of modern manufacturing needs the cooperation of multiple robots. Therefore, it is urgent to develop a group robot system with better compliance, consistency and optimization performance.

(3) adaptive neural network control of the manipulator

Multi-finger dexterous robot hand is a complex, dynamic coupling, nonlinear system with time-varying characteristics, such as system modeling error exists, high frequency characteristics, joint friction and signal detection error and so on the many kinds of uncertainty factors, the actual situation of the objective existence makes the control system performance is poor, so regular feedback technology can't meet the control requirements. Neural network has the characteristics of nonlinear transformation and highly parallel computing ability, which can effectively identify the parameters of the manipulator system, but it can not completely solve the uncertainty problems caused by the modeling error of the manipulator and external interference. In order to use neural network to carry out adaptive control of manipulator system, it is often necessary to integrate other algorithms, including sliding mode control, robust control and intelligent control. Currently studying reasoning based on knowledge rules and intelligent control algorithm, such as fuzzy control, learning control, expert control and genetic algorithm, particle swarm optimization algorithm, etc., have their advantage in processing system uncertainty, the method of multiple control combining organic combination, complement each other, and form a new control method has become more refers to the manipulator adaptive control in the field of research and development trend of the frontier.

(4) underwater autonomous navigation system

Autonomous underwater vehicles are widely used in underwater operations and are the basis of many scientific, industrial and military activities. Therefore, the realization of high-precision positioning of underwater vehicles and the collaborative navigation of multiple underwater vehicles has become the research frontier at home and abroad. Ultra-short baseline positioning system has been widely used in recent years. It is an underwater acoustic positioning technology with sound wave as the information carrier. The acoustic beacon of the underwater vehicle sends out the acoustic signal, and the ultra-short baseline array on the water surface receives the signal and calculates the underwater orientation and distance. The software scheme is provided by the algorithm based on kalman filter, extended kalman filter or decentralized extended information filter, and the hardware system consists of inertial measurement unit, fiber optic gyroscope, doppler log and other sensing elements. In the underwater positioning and navigation, the important factors that affect its precision is in view of the underwater vehicle motion estimation algorithm, this algorithm not only in the position error between the expected path and execution path ways to influence the outcome of positioning and navigation, but also affect the underwater vehicle for geographic reference data, so the autonomous underwater vehicle motion estimation algorithm requires accurate and light weight. In addition, as an alternative to the ultra-short baseline array, acoustic modems have great potential in acoustic navigation of single autonomous underwater vehicle (auv).

(5) li-ion battery thermal management technology
In recent years, under the dual threat of energy crisis and environmental pollution, the development of electric vehicles has attracted extensive attention. Lithium ion battery becomes the first choice for electric vehicle because of its advantages such as high energy density, strong power density, long cycle life and low self-discharge ratio. However, lithium ion battery will generate a lot of heat and may cause thermal escape in the process of rapid charging and discharging cycle with large current. If not properly handled, it will have a great impact on battery performance, service life, safety and other aspects. Experiments and numerical simulations are often used to analyze the thermal behavior of batteries and battery packs for better thermal management. Accurate thermal model is very important for the numerical simulation of lithium ion battery. It must accurately express the energy conservation, heat generation and boundary conditions of the battery. At present, the research methods of power cell cooling mainly focus on air cooling, liquid cooling, phase change material cooling, boiling cooling and heat pipe cooling. At the same time, the research of battery heating technology can not be ignored, which determines whether lithium ion battery can be used in low temperature and high altitude areas. All kinds of thermal management technologies have their own advantages and disadvantages. In practical application, cost, complexity, weight, cooling effect, temperature consistency and parasitic power consumption must be taken into account to achieve effective thermal management of lithium ion batteries.