Research On The Key Techniques Of Ultra-wideband Based Indoor Position And Orientation Estimation

As the popularization of the smartphone and the wearable device,location based services(LBS)become more and more popular,which are taken as an emerging industry to promote economic growth.As a result,indoor positioning technologies are spawned.Statistics show that one spends more than 80 percent of her/his time in the indoor environment.Therefore,how to obtain the indoor location information is a critical factor in the LBS.Today the emergence of 5G network,who owns high bandwidth and low delay,prepares a ground for the LBS and then for the indoor positioning technologies.According to the survey conducted by Marketsand Markets,the indoor positioning technologies will own a market of well over 40.99 billion dollars by 2022.Outdoors,Global Navigation Satellite System(GNSS)can be used for high-accurate positioning.However,indoors,GNSS will be failed to provide position estimation because its signal is attenuated even completely blocked by walls and roofs of the building.Consequently,various techniques are used for indoor location estimation,such as wireless local area network(WLAN)based,radio frequency identification(RFID)based,Bluetooth based,inertial measurement unit(IMU)based,vision based,ultrasound based,ultra-wideband(UWB)based,pseudolites based,and geomagnetism based.Therein,UWB is currently considered as a promising technology for high-accurate indoor positioning due to the advantage of high time resolution,good multipath resistance,a certain degree of obstacle penetration,and so on.Therefore,it has attracted the attention of researchers from both academy and industry.However,in the past decade,UWB based indoor positioning is limited to a particular group of applications,such as industry,logistics and robot localization and navigation,because of inadequate popularity of UWB chip.In September 2019,Apple announced that the latest release of i Phone(i.e.,i Phone 11)includes an UWB chip,which facilitates the popular of UWB chip in smartphones.As a result,it will become more and more continent to use UWB technology for indoor positioning.The history of UWB can be traced back to the spark gap transmission experiments performed by Marconi and Hertz in the late 1890 s.However,it is limited to military applications until the Federal Communications Commissions(FCC)in the USA allowed unlicensed use of UWB products in 2002.Thereafter,this technology gained a huge boost and was gradually used to indoor position-ing.Until now,some key problems in UWB based indoor positioning have not been well solved,such as NLOS identification and mitigation.Also,the recent UWB based indoor positioning systems only provide position estimation without orientation information.In addition,in the recent years,as the increase of the type of built-in sensors and the computing power of portable electronic devices,integrated positioning based on multiple sensors have become a hot topic in indoor positioning.However,to the best of my knowledge,there is currently very little research on the integrated positioning based on UWB and other sensors.Therefore,the dissertation focuses on some key problems in UWB based indoor positioning,including NLOS identification and mitigation,UWB and IMU based integrated positioning,and low-cost orientation estimation in indoor environments.The main contributions of this dissertation are summarized as follows:(1)A fuzzy theory based NLOS identification and mitigation method is proposed.The current NLOS identification and mitigation methods generally require a priori knowledge,is limited to specific scenarios and have a high computation complexity.The proposed method classifies the indoor signal channels into few categories and then identify the NLOS ranges and correct them based on fuzzy comprehensive evaluation,which is independent of a prior knowledge and is suitable for different scenarios.(2)In order to further mitigate the influence of the residual NLOS range error on location estimation,a method is proposed to select an appropriate range set for location estimation.Meanwhile,the roust estimation theory is introduced into the location estimation algorithm.(3)UWB/IMU loosely-coupled and tightly-coupled systems are designed.The observability of the designed systems are analyzed.Meanwhile,the influence of the number of UWB anchors and their geometry on UWB/IMU tightly-coupled system is also analyzed.In addition,a zero acceleration update(ZAUP)based horizontal attitude estimation algorithm for pedestrian localization is proposed.The algorithm be independent of the assumption of small attitude error.Meanwhile,the algorithm can be implemented by a direct state KF.(4)An orientation estimation algorithm based on range measurements of multi-antenna UWB is proposed.The indoor orientation estimation is currently depend on the called attitude and heading reference system(AHRS).Such an orientation estimation method is easily effected by indoor magnetic disturbances.The proposed method has no need of geomagnetic fields and thus is more practical indoors.