Research On Anti-jamming Technology For 6-DoF Ultrasonic Positioning System

With the development of technology and the continuous improvement of people’s living standards,lightweight and highly immersive virtual reality devices are an inevitable trend in technological development.Six-degrees-of-freedom(6-Do F)positioning technology is the core technology of virtual reality devices,with hardware simplicity as the foundation for lightweight devices.Real-time performance,high positioning accuracy,and high anti-interference capability are necessary conditions for achieving high immersion.Compared with optical-based positioning technology,ultrasound positioning has the advantages of low power consumption,simple structure,high positioning accuracy,and strong real-time performance,and is one of the current solutions for virtual reality positioning.The field of virtual reality demands higher requirements for the precision,real-time performance,and anti-interference capability of 6-Do F positioning.Currently,commercial virtual reality devices are mostly based on optical 6-Do F positioning technology.However,ultrasound positioning has the advantage of being structurally simple and easy to miniaturize and lightweight compared to optical positioning.Additionally,due to the high spatial resolution of ultrasound,it can achieve high-precision positioning at close range.However,in the field of virtual reality,the problem of interference with 6-Do F ultrasound positioning has not yet been solved.This article first introduces the 6-Do F ultrasound positioning method,and then introduces the inertial measurement unit to address the problem of interference from samefrequency and occlusion in ultrasound positioning.Two techniques are proposed:ultrasound anti-same-frequency interference based on a filtering model and ultrasound anti-occlusion interference based on a deep model.Compared to the filtering model,the deep model takes ultrasound coordinates as input to ensure that even if some ultrasound receivers are occluded,and the ultrasound coordinates are insufficient to calculate the attitude,the remaining ultrasound coordinates can still provide correction for the inertial measurement unit to calculate the attitude,ensuring the system’s anti-occlusion interference capability.The main contents of this article are as follows:(1)A 6-Do F ultrasound positioning solution with three transmitters and three receivers is proposed to address the implementation problem of 6-Do F ultrasound positioning in the field of virtual reality.The system employs a time-sharing transmission and reception mechanism to reduce the impact of same-frequency interference.Based on the principle of three points determining a plane,three ultrasound transmitters and three ultrasound receivers are set up to form two planes.The rotation relationship between the two planes is solved to obtain the quaternion description of the target attitude in the base station coordinate system,thereby achieving 6-Do F ultrasound positioning.(2)In order to address the issue of same-frequency interference in ultrasound localization,an inertial measurement unit is introduced and a multi-filter method based ultrasound anti-same-frequency interference technology is proposed.This technology achieves a refresh rate of up to 80 Hz and an average positioning accuracy of centimeterlevel in a specific direction,with an average attitude error of about 5 degrees in six degrees of freedom localization in complex environments.The method fuses the target attitude output by the inertial measurement unit with the target attitude output by the six degrees of freedom ultrasound localization method through an error state Kalman filter,greatly improving the system’s anti-same-frequency interference capability.At the same time,when ultrasound is subjected to occlusion interference,the inertial measurement unit can provide temporary real-time target attitude information.(3)To address the problem of six degrees of freedom ultrasound localization failure caused by occlusion interference,a tightly coupled Transformer model is introduced,which is sensitive to time-series data.Using the data from three ultrasound receivers’ coordinates and the inertial measurement unit as input,the model ensures that when one ultrasound receiver is occluded,the coordinate data from other ultrasound receivers can correct for the inertial measurement unit,achieving a refresh rate of 200 Hz and centimeter-level positioning accuracy,while keeping the average attitude error within 5degrees.When an ultrasound receiver is occluded,the filtering model degenerates into two isolated systems: ultrasound localization and inertial measurement unit orientation,resulting in a sharp drop in attitude accuracy.Therefore,we propose a tightly coupled Transformer model and divide the deep model into a single encoder deep model with a refresh rate of up to 1000 Hz and a multi-encoder deep model with higher six degrees of freedom positioning accuracy and anti-interference properties to meet different scene requirements.