Design And Development Of The Imaging Bionic Polarized Navigation Device

With the coming of information era, navigation technology plays more and more important role in in the people’s living and the military applications. Compared with the inertial navigation and satellite navigation, celestial navigation has the advantages of strong autonomy, good concealment, high reliability and no error accumulation. But its complex structure, high cost and large volume limits the range of its application.Through experiments, modern biologists found that through experiments, dung beetle, ant and other insects can use its polarization vision (visual organization, structure and related nervous system which is extremely sensitive to the polarization direction of skylight) to induce the skylight polarization distribution, and analyze the polarization azimuth in the brain to navigate in motion. The bionic navigation based on skylight is interfered less by the outside environment and has no error increasing with time gradually. This method has broad prospects for development.In this paper, the imaging bionic polarized navigation device imitate insect polarization vision, utilize commercial CMOS image sensor and DSP combined with micro-nano processing technology, and could calculate the angle between zero reference line and meridian based on the skylight at fixed time and specific locations and then output the information of the heading angle, and better adaptability and robustness were performed.The contents of this paper are mainly showed in the following aspects:Firstly, this paper studied the sky light polarization distribution model, analyzes the working principle of imaging bionic navigation device, designed the architecture imaging bionic navigation device and built the hardware system of the device. The device mainly consists of a telephoto lens, plastic polarizer or metal polarizer, imaging sensor and digital signal processor.Secondly, the program of the embedded system in the device was designed and compiled, to achieve the functions of polarization image acquisition, data processing and polarization angle calculation.Thirdly, this paper studied a polarization algorithm, extracted intensity values of pixels in homogeneous regions, eliminated noise contained in image and compensate errors which occur in the product of nanowire polarization filters, solved the overdetermined system of linear equations, and gave a set of least square solutions of the Stokes parameters and the angle of polarization as output to improve the accuracy and robustness of the device.Fourthly, the performance of the device inside and outside was tested and calibrated.