| The level of facilities agricultural mechanization is relatively low at present in our country.Generally there are problem of low production efficiency,high labor intensity and need more labor and so on in facilities agriculture.Along with the agricultural labor shortages and the aging of the population,the price of rural labor force is higher and higher,so we must accelerate mechanization and automation of the facility agriculture,and make mechanization and automation of production way to replace manual labor,thus to further improve the production efficiency and benefit of facilities agriculture.So that facility agriculture robot will have broad prospect of research and application which is combination of mechanization,automation and intelligence.The automatic navigation and positioning is one of the key technologies of facility agriculture robot,widely used in facility agricultural production of spraying,fertilizing,weeding,harvesting and transport,etc.Although navigation research of facility agricultural machinery has been a lot of progress in our country,the price of high precision of navigation system is very expensive,which is unfavorable for extension and application in rural China.Therefore the development of high precision,low cost of facilities agricultural machinery automatic navigation and positioning system is an urgent problem need to be solved.Considering the working environment and application needs of facility agriculture robot navigation and positioning system,in order to reduce cost and improve the accuracy and reliability of the navigation system,the facility agriculture robot navigation and positioning system are studied in this paper.The relevant theoretical research and experimental analysis has carried on,such as localization algorithm of facility agriculture robot based on wireless sensor network(WSN),error sources and calibration method of navigation sensor in the integrated navigation system,estimation algorithm of attitude angle based on Kalman filter and multi-sensor information fusion,the composition,realization and application of the facilities agriculture robot integrated navigation system and so on.The main research contents of this paper are as follows:(1)Research on WSN location algorithm based on Kalman filter and time of arrival(TOA).The theoretical analysis and comparative study of WSN localization based on received signal strength indicator(RSSI)and TOA are carried out.And a set of WSN positioning system with high precision and low cost is designed for the facility agricultural robot.The TOA-based WSN localization algorithm with low hardware cost was studied mainly.The distances between unknown nodes and each beacon node were calculated by using Symmetrical double-sided two ways ranging(SDSTWR)algorithm which was based on chirp spread spectrum(CSS)technology in location node.And then the coordinate information of the unknown node was calculated by the triangle centroid localization algorithm.Finally in order to resolve the effects of measurement noise on positioning accuracy,the iterating based on Kalman filter algorithm was applied to observed values of target state.The results of positioning experiments indicate that WSN nodes based on TOA ranging accuracy is high,and the stability of WSN positioning system based on Kalman filter and triangle centroid localization algorithm is good and the positioning accuracy is high.And the positioning precision could reach about 1 m within the scope of 3 edges of equilateral triangle formed by 3 beacon nodes.(2)The hardware composition design of the low cost integrated navigation system is studied.And the error analysis and calibration method of the sensor in the integrated navigation system are studied.The hardware composition of the integrated navigation system for facility agricultural robots is designed.The error sources,calibration principle and calibration method of magnetoresistive electronic compass,triaxial MEMS gyroscope and triaxial MEMS accelerometer in the integrated navigation system are analyzed theoretically and experimental verification.A calibration method combining the ellipse hypothesis compensation method and the least square method to calculate the compensation coefficient is put forward.From the experimental data we can see that the calibration method is simple,feasible,correct and effective.We designed the scheme of calibration experiments,dynamic and swing experiments of the three-axis MEMS gyroscope,which could provide the reference for the calibration and performance test of all three-axis MEMS gyroscopes.(3)Research on attitude estimation algorithm of low cost INS based on Kalman filter and multi-sensor information fusion.It mainly studies the modified quaternion full attitude algorithm based on Kalman filter and multi-sensor information fusion,which uses accelerometer and magnetometer to modify the gyroscope.The algorithm is applied to the estimation of the attitude angle of the low cost INS.The experimental results show that the algorithm can effectively reduce the signal noise and respond to the movement of the carrier in time,and it can provide the navigation system with more accurate attitude angle information.(4)Platform construction,implementation and application of integrated navigation system of facility agricultural robot are researched.A tracked robot is used as the walking chassis of the facility agricultural robot,and an integrated navigation system based on low-cost INS,visual navigation subsystem,electronic compass and WSN positioning system is built on the chassis.The modularized software programming design is adopted based on the hardware platform of the facility agricultural robot integrated navigation system.The autonomous navigation control function of the integrated navigation system is realized by fuzzy control and PID control algorithm.The navigation application of the integrated navigation system of the facility agricultural robot is studied in the indoor environment and the greenhouse environment.The experimental results show that in the integrated navigation system,the visual navigation system has high precision and good adaptability.And the navigation accuracy of low cost INS based on Kalman filter and multi-sensor information fusion is also high in short time.However,the navigation accuracy of electronic compass navigation is relatively low,but it is found by experiment that the electronic compass is suitable for the detection of the yaw angle when the facility agricultural robot is turning around. |
PDF Full Text Request