Abstract
We present a short-range magnetic positioning system that can track in real-time both the position and attitude (i.e., the orientation of the principal axes of an object in space) of up to six moving nodes. Moving nodes are small solenoids coupled with a capacitor (resonant circuit) and supplied with an oscillating voltage. Active moving nodes are detected by measuring the voltage that they induce on a three-dimensional matrix of passive coils. Data on each receiving coil are acquired simultaneously by a distributed data-acquisition architecture. Then, they are sent to a computer that calculates the position and attitude of each moving node. The entire process is run in real-time: the system can perform 62 position and attitude measurements per second when tracking six nodes simultaneously and up to 124 measurements per second when tracking one node only. Different active nodes are identified using a frequency-division multiple access technique. The position and angular resolution of the system have been experimentally estimated by tracking active nodes along a reference trajectory traced by a robotic arm. The factors limiting the viability of upscaling the system with more than six active nodes are discussed.
Highlights
Magnetic positioning systems (MPS) are widely investigated as a suitable solution for applications needing localization in space of magnetic transmitters aptly mounted on objects or people [1].The growing field of the Internet of Things (IoT) provides many application scenarios for localization systems [2], such as wireless sensor networks [3,4], mobile robots [5], and location-based services [6].The main advantage of using localization systems based on a magnetic field is that they do not need line-of-sight conditions to work
The Texas Instrument (TI) Delfino includes a floating point unit (FPU) co-processor for each core, supporting 32-bit single-precision floating point operations, that we used with the specific mathematical library provided by TI implementing a device-optimized FFT algorithm [37]; an interprocessor communication (IPC) module for communications between the two cores, that we used to control parallel operations; and a pulse width modulator (PWM), that we used to generate a square wave applied to trigger the analog-to-digital converters (ADC) at fixed sampling frequency
The first interesting result is that, they are operated using different frequencies, the accuracy is practically the same for each TX, i.e., varying the frequency within the band defined in Section 2.8, with f c = 182 kHz and B = 10 kHz, does not produce any appreciable difference in the accuracy of the system
Summary
Magnetic positioning systems (MPS) are widely investigated as a suitable solution for applications needing localization in space of magnetic transmitters aptly mounted on objects or people [1]. The main advantage of using localization systems based on a magnetic field is that they do not need line-of-sight conditions to work For this reason, an extensively investigated application is the tracking of capsule probes inside the human body, i.e., wireless capsule endoscopy [7,8]. Short-range MPSs [12,13] can be used to accurately navigate surgical instruments [13,14], to investigate motor symptoms of diseases such as Parkinson’s disease [15,16], to provide a human–machine interface such as in data gloves [17,18] In the latter case, magnetic transmitters and/or sensors are mounted on the fingers in order to track their movements. Multi-node tracking can be exploited in data gloves equipped with magnetic sensors, e.g., to assess hand kinematics and functioning [18,26], to drive industrial equipments by gesture recognition [27], or to devise automatic sign-language recognition systems [28]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
