Indoor positioning floor based on pressure sensing touch technology

introduction

Indoor Positioning System (IPS) is an important part of ubiquitous computing. With the popularization of smart terminals and the advent of the mobile Internet era, the demand for positioning and navigation in complex indoor environments is increasing. At the same time, IPS has been widely used in public security, mobile e-commerce and other fields. At present, the positioning technologies used by IPS mainly include RFID, Bluetooth, WiFi, UWB, ZigBee, optics, ultrasound, infrared, geomagnetic field, etc. Indoor positioning systems that use a single positioning technology have certain defects, such as high burden, poor scalability and robustness, low positioning accuracy, and long response time. Therefore, the integration of multiple positioning technologies is a development direction of IPS.

Force Based Touch Sereen Technology zTouch comes from F-Origin, an American company. The principle of this technology is to set up multiple pressure sensors at the corners of the rigid panel. The sensors sense the pressure perpendicular to the panel generated by the user's touch or gesture on the panel, and locate the sensor according to the different forces on each point. This technology has the advantages of good environmental robustness, strong durability, no burden, three-dimensional information perception, and high positioning accuracy. Its application in IPS can make up for the shortcomings of other positioning technologies. At present, many foreign institutions are studying positioning systems based on this technology, such as the Weight Lab of Lancaster University in the UK, Sheffield Hallam University in the UK, and Smart Floor of RWTH Aachen University in Germany.

This paper designs a smart floor based on pressure sensing touch technology, and uses the floor to build a small positioning system. The system uses 16 pressure sensors and 4 floors to build a positioning platform, which can realize the positioning and path display of the moving car on the platform, and can view the car's position information through LabVIEW software and Android terminal.

1 Positioning principle

The basic unit of pressure sensing touch positioning technology consists of a rigid panel and four pressure sensors. The pressure sensors are distributed at the four corners of the rigid panel to form a four-point structure. The four-point positioning model is shown in Figure 1.

In the model shown in Figure 1, Si (i=1, 2, 3, 4) represents the i-th pressure sensor. In the two-dimensional coordinate system of the P plane, the coordinates of the contact point (or the center point of the contact surface) between Si and the P plane are recorded as (xi, yi), and the rectangular area (in the dotted box) formed by the four contact points is called the positioning area D. Apply a force F perpendicular to the P plane at any point (x, y) in D, and let Fi represent the partial pressure of F on Si. According to the calculation formula of the center of gravity position of a four-point plane object, equations (1) to (3) are established.

According to equations (1) to (3), the position of force F can be calculated, thereby achieving positioning.

2 System Structure Design

2.1 System Design

According to the experimental conditions of this study, a positioning demonstration system for a small car was built. The system consists of a positioning platform, a positioning object (small car), a data acquisition card, a PC, and an Android monitoring terminal. When the positioning object is stationary or moving on the positioning platform, its absolute position coordinates (X, Y) in the physical space are calculated by analyzing the data of the sensor. The position attributes are mapped and reflected in the virtual map of the upper computer position management server (LabVIEW software). The position management server can build a virtual scene and provide location query and navigation services for mobile terminals. The positioning system composition is shown in Figure 2.

This system uses USB-6259 (BNC) to collect data from the positioning platform sensor. USB-6259 (BNC) is a high-speed, multi-function M series data acquisition card produced by NI Corporation of the United States. It provides 16 analog input channels, with a maximum sampling rate of 1.25 Ms/s and a sampling accuracy of 16 bits, which meets the system requirements.

The positioning system can be divided into 3 modules:

① Positioning display module. This module is implemented on the LabVIEW platform of the PC to display the absolute position and motion path of the car on the positioning platform in real time.

② Motion control module. This module controls the movement of the car through the LabVIEW program and transmits control instructions through Bluetooth to achieve the purpose of non-contact movement of the car, while preventing the lateral force generated by the contact method from affecting the positioning accuracy.

③ Remote monitoring module. This module enables remote monitoring of the car's location through an Android phone, and the PC broadcasts the car's location information to the remote terminal through the Internet.

2.2 Positioning platform settings

The positioning platform consists of 16 sensors and 4 glass plates, with a total area of ​​70×70 cm2, divided into 4 units. The sensor adopts a spoke-type pressure sensor (model BK-4). The sensor itself has an integrated bridge and has 4 external leads, namely power supply Vbg, GND, signal + and signal -. Since the signal output by the sensor is weak (μV~mV level), the signal needs to be amplified. The amplifier used in this system is AD623.

To ensure accurate positioning, the positioning platform needs to be leveled. When the positioning platform is not level, that is, when the sensor heights are inconsistent, one or two pivots in the four-pivot positioning model shown in Figure 1 may be in a suspended state without force, which is the "virtual leg" phenomenon. The "virtual leg" phenomenon will change the distribution of the force on the four-pivot platform, so the four-pivot formula cannot be used to calculate the center of gravity. Zhou Zulian proved through a mathematical model that the four pivots in the four-pivot structure can be stressed at the same time, that is, the "virtual leg" phenomenon can be avoided by leveling.

The simple leveling of the four-point platform usually adopts the method of moving closer to the highest support, that is, adjusting the three lowest heights of the four points through shims to make them consistent with the height of the highest point to adjust the platform surface. This system levels the positioning platform by padding paper under the sensor.

3 System Software Design

The host computer development platform of this system adopts LabVIEW 2013 launched by NI. The designed software interface is shown in Figure 3.

3.1 Positioning display module

The input of the positioning display module is the data of 16 pressure sensors. After filtering and calculation, the output is the position of the center of gravity of the object on the positioning platform. The position display mode includes text coordinates and two-dimensional pictures, and the path display mode is an XY graph. The positioning display process is shown in Figure 4.

3.2 Motion Control Module

The motion control module is realized by establishing Bluetooth communication between the PC and the car. The PC without Bluetooth function can realize Bluetooth function through Bluetooth adapter and BlueSoleil software. The car end is connected to a Bluetooth transparent transmission module through the UART interface. At this time, the PC end and the car end establish a virtual serial port connection. The LabVIEW end sends motion instructions to the car through VISA serial port communication programming. This module can complete the speed control of the car and the control of five motion states: forward, backward, left turn, right turn, and stop.

3.3 Remote Monitoring Module

In the remote monitoring module, the PC is the location management server, the Android terminal is the client, and the network architecture adopts a distributed Publisher-Subscriber model. The PC provides location query services for all registered Android terminals.

The remote monitoring module uses the cross-platform network communication plug-in SCCT. SCCT (Smart Phone Cross-Plat-form Communication Toolkit) is a development kit developed by T4SM (Tools for Smart Minds Software Solutions) that supports multiple platforms and programming languages. It is used for data communication between multiple platforms and the LabVIEW platform. SCCT supports operating systems including Windows, Linux, iOS, Android, and Phone7, and programming languages ​​including LabVIEW, Java, C, Object C, .NET, and Javascript (HTML5). In this system, the LabVIEW side uses the VI provided by the SCCTPublisher Library to implement the functions of verifying the identity of the subscriber requesting the connection, checking the connection status, broadcasting data to all active subscribers, and receiving subscriber requests. The Android side uses the API provided by the SCCT Subscriber Library to implement functions such as sending requests to the publisher, receiving data, and submitting it to the interface for display.

The flowchart of the information publishing program on the LabVIEW side is shown in Figure 5. The software architecture on the Android side is shown in Figure 6.

4 System Testing and Result Analysis

The system test scenario is shown in Figure 7.

The remote monitoring interface of the Android phone is shown in Figure 8.

After testing, this system has realized the functions of positioning display, motion control and remote monitoring of the car.

In the positioning accuracy test, a position reference grid is drawn on the platform with a grid side length of 1 cm. 10 test points are randomly selected on the positioning platform, pressure is applied at each point, and measurement is performed 6 times. The coordinates calculated by the software are recorded and compared with the actual grid coordinates. In the data analysis process, the following concepts are defined: