Research on swarm intelligence collaborative work and cognitive computing technology

   Hello everyone , I'm the third person in charge of the project , Chao Qing . Representative topic 3 , Participants in the research of swarm intelligence collaborative operation and cognitive computing technology present to experts Report the relevant contents of Topic 3 . The main participants of Topic 3 include ： Tsinghua University , Heilongjiang devo Technology Development Co., Ltd .

   The content of the project will be reported in the following five parts .

   Firstly, it introduces the background and main research contents of the project .

   project 《 Swarm intelligence, autonomous operation, smart farm 》 Based on the current situation of agriculture in China , Implement the strategy of precision agriculture . Aiming at many difficulties in modern agricultural industry , Apply a new generation of artificial intelligence technology , Meet the needs of upgrading the agricultural industry .

   The research content of swarm intelligence cooperative operation and cognitive computing in topic 3 includes three technologies ： 1） Based on the characteristics of agricultural operations , That is, under the non structural complex farmland environment , Autonomous decision-making and task allocation of agricultural machinery 、 Man machine integration technology ; 2） Efficient and reliable on-site wireless real-time communication and embedded edge cloud computing platform technology ; 3） In the process of cultivation, there are many ways to manage the harvest 、 Autonomous driving of multiple sets of agricultural machinery and efficient cooperative operation technology of formation ;

   Combine the specific crop varieties produced by smart farm ： Wheat 、 Practical solutions proposed in corn and rice production ： （1） Autonomous decision-making and task allocation in swarm intelligence ; Formation motion control based on reinforcement learning ; （2） Complete the technical breakthrough in man-machine cooperation based on intention recognition ; Develop an embedded computing platform to meet the project demonstration and verification , Realize on-site real-time wireless networking and intelligent decision planning algorithm ; （3） Finally, the agricultural machinery group in all links of farming, management and collection will work together efficiently and intelligently in the whole production process ;

   It shows the relationship between Topic 3 and other topics , And the internal main research content structure . Task 4 assigns cluster tasks to task 3 ; Topic 3 is on the self-developed real-time communication and edge computing platform , Complete the internal independent decision-making algorithm , Achieve site formation control , Real time feedback between man and machine is realized through man-machine cooperation interface . The specific real-time instruction is sent to the agricultural machinery interface in task 2 . Cooperate with the overall completion of agricultural intellectualization 、 Preparation of industry standards for autonomous and unmanned operation .

  （2） Main technical routes and technological innovation .

   In terms of autonomous decision-making and task allocation of agricultural machinery groups , Aiming at the complex dynamic environment of farmland operation , Use the knowledge base established by other topics of the project , Design and develop autonomous decision-making and task allocation algorithms . Mainly based on evolutionary game algorithm 、 Bionic cluster decision algorithm and swarm intelligence optimization algorithm realize group task allocation and scheduling .

   In the agricultural machinery group formation In terms of cooperation with . On the one hand, the travel characteristics of different agricultural machinery are very different , On the other hand, farmland sites have certain characteristics of uncertainty and unstructured . Use reinforcement learning , Realize the optimization of formation control parameters and strategies , And real-time control .

   Human computer interaction is to complete all stages of intelligent field operation of agricultural machinery group , You know 、 controllable 、 An important guarantee of analyzability . Determine the safe and reliable boundary conditions of unmanned system . On the one hand, the operation monitoring personnel can understand the operation status of agricultural machinery through handheld or wearing electronic equipment ; On the other hand, use sound 、 Images 、 Gesture recognition technology completes the human-computer cooperation of intention recognition .

   Aiming at the operation task and environment of farm tools on site , Develop a wireless network that can be used on site 、 Hardware platform for embedded computing . Complete the calculation of autonomous planning tasks , And send the real-time control command to the lower machine control unit of agricultural machinery , And complete the collection of job information and operation status .

   The main innovations of the project include two aspects ： 1） A dual drive strategy based on big data and knowledge map is proposed , Combining artificial intelligence means such as bionics and intensive training , A method for realizing the collaborative operation planning of heterogeneous agricultural machines and tools and swarm agent ; 2） Propose and implement Including man-machine integration and cooperation 、 Dynamic formation of agricultural machinery with large-scale real-time communication and embedded computing 、 Collaborative work mode ;

  （3） Research team and work foundation

   The lead unit of Topic 3 , Tsinghua University relies on the Institute of navigation and control, Department of automation, Tsinghua University . It is mainly based on the core technology of intelligent unmanned system , In UAV 、 Unmanned vehicle 、 Space robots 、 Underwater Unmanned system has undertaken and completed a number of national projects 、 Ministry of Science and Technology 、 Scientific research projects such as NSFC ; Participant: Heilongjiang zhiwo Technology Development Co., Ltd , It is a national high-tech enterprise specializing in building high-end agricultural machinery and agricultural services . Has a number of patented technologies , Won many scientific and Technological Awards .

   Here are some on-site scientific research projects of participating units , And related agricultural machines and tools .

   In terms of previous technology accumulation , The communication networking system of high dynamic UAV group is realized , It can meet the on-site communication requirements of group agricultural machinery ; The distributed cooperative control in biological population is proposed and verified And optimization mechanism , And realized a variety of unmanned and mission planning Algorithms related to collaborative control .

   In the task application 、 During startup , Carried out paddy field research together with other subjects of the project 、 Liu Sheng's research on the standardized operation of dryland agricultural machinery is to sort out ; The autonomous operation scheme of agricultural machinery group demonstrated by the overall project is determined ;

   The demonstration base for the application of the subject achievements was investigated on site ： 290 Farm site network 、 Power and farmland facilities .

  （4） Project schedule and assessment method

   This table gives three years , Schedule by quarter . Topic 3 is carried out in parallel according to three main lines ; The first one is about the simulation of agricultural machinery cooperative operation algorithm Design and verification platform ; The second is the verification design and finalization of the computing hardware platform to realize swarm intelligent control ; The third is the report generated from the summary of relevant technologies in the development process 、 The paper 、 patent 、 Industry norms and talent training . Finally, cooperate with the project to complete the system integration and application demonstration .

   About the assessment method of the project, the contents of each part are concentrated in this form .

  （5） Project budget and expected results

   This table shows the total funds of the project and the budget allocation plan .

   The expected results include three aspects ： The first aspect is the algorithm software and hardware platform of swarm intelligence ; The second part is the realization of system integration and demonstration , Including independent decision-making of more than three kinds of unmanned agricultural machinery And task assignment ; Realization 5 Above the platform set, feel the same link, independent formation and cooperative operation ; The third part is related papers 、 patent 、 Soft content . Finally, three items of agricultural intellectualization will be formed 、 Independent and unmanned industry standards .

   The above is my report , Please comment and correct by all experts .

---