Research On Position And Orientation Constraints Based Bio-inspired Navigation

True autonomous navigation technology of long duration and high-precision becomes one of the main factors that restrict the development of autonomous unmanned system.Nowadays,the navigation of unmanned system mainly depends on the satellite navigation system.However,the satellite navigation signal is easy to be disturbed,and it may be unavailable in the wartime.Inertial navigation system(INS)has been core navigation equipment for unmanned system because of its excellences such as independence and whole navigation information.However,due to the existing shortcomings that the navigation error increases with time and lacking long term stability,the INS cannot meet the long duration navigation mission.With recent progressed in the technology of bionics,microelectronics and micro nano,the bio-inspired navigation technology has become a research hot point of navigation.In order to meet the needs of the autonomous navigation technology of the autonomous unmanned system,this thesis mainly focuses on the mechanism of animal navigation,the location recognition algorithm based on grid cells and place cells,the bio-inspired polarization navigation method,and bio-inspired navigation method in the hybrid space based on multi-sensor fusion technology.Simulations and practical tests have been carried out to verify the effectiveness of the proposed algorithms.The main contributions and innovation points are summarized as follows:(1)Aiming at the problem of misrecognition and hard calculation of the existent location recognition algorithms,a bio-inspired location recognition method based on the activation mechanism of both grid cells and place cells is proposed.The accuracy and computational efficiency improvement compared to the existent algorithms is verified by using experiments.(2)Comparing to the single-scattering Rayleigh model,the Mie scattering model is more consistent with the polarization patterns of skylight.However,the theoretical studies and experimental results reveal that the single-scattering Rayleigh model is the only model which can be utilized for polarization navigation.The skylight polarization patterns in different atmospheric conditions are compared with the patterns calculated by the single-scattering Rayleigh model and these differences are quantified.The relationship between the degree of polarization(DOP)and the deviation of the angle of polarization(AOP)is thoroughly studied.These research work provide a solid theoretical and technical support for the orientation technology utilizing the skylight polarization patterns.(3)Utilizing all the outputs of the polarization navigation sensor,a new compass information extracting method based on the least-squares algorithm is proposed.The results of simulation and experiments demonstrate the efficiency and accuracy of the proposed algorithm.A new calibration model by formulating the calibration problem as a multi-objective optimization problem is presented.Unlike existing calibration models,the proposed model makes the calibration problem well-posed.The calibration parameters are optimized through Non-dominated Sorting Genetic Algorithm-II(NSGA-II)approach to minimize both AOP residuals and DOP dispersions.The results of simulation and experiments show that the proposed algorithm is more stable than the compared method for the calibration application of polarization navigation sensors.And then,a heading angle determination method utilizing DOP and horizontal angle information is put forward.The road tests results show that the root mean square of the heading angle errors is reduced.(4)Aiming at the problem of visual navigation error increases with time and lacking long term stability,a polarization skylight and vision integration navigation method in Euclidean space is presented.On the basis that topological and Euclidean space navigation methods have complementary strengths,a hybrid space navigation method that combines their strengths and avoids their weaknesses is presented.The topological map is built based on the grid cells activation mechanism,and the location recognition is fulfilled based on the place cells activation mechanism.The bio-inspired hybrid space navigation method effectively suppresses the error increase of both the heading angle and positioning.The experiment results provide the preliminary feasibility of the novel bio-inspired navigation technique for the long duration and high-precision application of autonomous unmanned system.