USING THE SYNERGY METHOD FOR IMPROVEMENT OF THE ACCURACY OF LOCATION BASED AUGMENTED REALITY

The key components of a location-based augmented reality application are data collected from various sensors and methods for processing these data. However, the accuracy of the sensors does not always meets the requirements, processing the collected data require significant computational resources that have a negative effect to the power consumption and overall performance of the application. The article considers possible ways of solving these problems by using various data analyses and processing methods. One of the largest segments of augmented reality market is mobile software segment, which is mainly represented by software for smartphones. At the same time, most of the used devices have disadvantages deriving from their small size and mobility. These are: limited charge of a portable power source, low performance in comparison with desktop computers, inefficient cooling system and limited set of built-in sensors. Augmented reality applications are focused on visualization of supplemented data at the right time and in the right place. Location-based augmented reality relates to technologies that use the device capabilities to determine its spatial position based on GNSS data, as well as compass, gyro, depth camera and other sensors displaying the augmented reality around. The need to research and analyse the data obtained from smartphone sensors is caused by the fact that in the location-based augmented reality application, developed by the authors, there were problems with the accuracy and stability of the visualization of 3D objects in virtual space. Data, obtained from the smartphone sensors (compass, gyro, accelerometer, GPS / GLONASS), are used to position the virtual camera in a 3D scene with information tabs. Since the visual "jitter" of 3D objects is significant, it was necessary to find out which of the sensors deals the greatest error. Due to the lack of ready-made solutions to create location-based AR systems, the authors met problems associated with extremely high power consumption, accuracy and stability of visualization of augmented reality elements, and need of intellectual use of smartphone sensors, disabling those that can be ignored. Unity software, which has all necessary tools for obtaining data from smartphone sensors, was used to analyze the data received from the smartphone, because it Based on the analysis of the data received from various smartphone sensors, a number of recommendations on how to work with these data and use algorithms to smooth and reduce noise within the creation of location-based augmented reality application was made.