Indoor location and navigation

CPSACPSA2

Aplicaziones presents in the Spanish market a localization solution, in interiors and exteriors, of people and things.

It is based on the developments made by our company during the last years and by the incorporation of alliances with third parties to give a complete solution.

Our company has been working on this type of solutions for almost 10 years. 

We started as an integrator of localization solutions based on WiFi technologies. The solutions worked reasonably, but at that time the precision was quite bad (1/2 the distance between access points), the position in the Z axis was not well resolved and mainly the applications were oriented to the location of computer assets. In addition, the response times were not very good and the costs were important (sometimes more than the WiFi deployment itself). All these drawbacks made this type of solutions have limited success in the market.

However, it was clear that there was a lawsuit..
So, we decided to organize an R&D strategy to explore the future of indoor location systems and if necessary develop an own IP.
In this way we began to contact companies and research centers pioneers in this type of solutions. As a result of this work we participated in several European programmes (H2020, Eurostars and Einnovest). Being awarded in some of them.

Locating people indoors

Our proposal is based on using an intelligent device that extracts the location of the person based on their movements (e.g. pedometer) and without the need to use external reference infrastructure.
In fact, this type of navigation has been used for many years by ships and airplanes, but at that time (6 years ago), the challenges for it to be used for locating people were huge.

To achieve this, specialized algorithms had to be developed and microelectronics had to be expected to reduce costs. It also helped that the military industry was interested in studying such solutions for the soldier programs of the future. Here is a presentation of the first specialized HW-based solutions we made for the Army's ITM (Marañosa Institute of Technology).

These devices used multiple sensors (accelerometers, barometers, magnetometers, and gyroscopes) located on the shoe or belt to estimate the person's location.
They were very expensive equipment and had little success because most potential customers were not willing to pay for them.
However, everything changed with the emergence of smartphones.
Suddenly users had a device with more than enough processing capacity to run the algorithms, so the cost of such solutions was substantially reduced.
However, there were some problems to be solved:

1. Smartphone sensors are low cost and therefore their accuracy is low.
2. The battery suffered with calculations for location and its duration was small (1 or 2 hours)
So we're looking for strategies to solve these problems.
a) On the one hand, algorithms with reference systems were incorporated to correct the drift in inertial navigation of smartphone sensors.
b) On the other hand, part of the location calculations were passed to the cloud.
This multiple strategy used by our solution is represented in a graph below:

 Fusion strategies

We see that we use GNSS (GPS or now Galileo). In general this is usually of little use indoors, but we hope that something will change in the future with the appearance of new multifrequency chips in smartphones.
We also run inertial navigation algorithms using smartphone sensors.
Finally, and to correct the drift of these sensors, we perform a mapping of magnetic fields and radio signals of the scenario (what is often called opportunity signals). We then compare this stored information, which is obtained in real time by the device to help with position determination.

In this blog post you can see the first results of this R&D.

As the smartphone is a very versatile device, with location information, we can also recommend routes to users. For example, in emergency situations.

 Rutas

Examples of some projects carried out with this solution can be found in these videos:

 

Locating things indoors

RFID has been a widely deployed indoor asset management solution for many years.
There are two cases of use of this technology. The solutions based on passive tags and those based on active tags are detected when passing through arcs at a small distance (e.g. 0.5m).
Passive tags are widely used in multiple applications, being those of the distribution market the best known.
The active ones are used to control high value assets and are based on deploying multiple radio equipment (receivers) around the scenario. The assets are then associated with an active tag (with battery). The location is done by detecting this tag from several receiver devices. In general, algorithms such as "RSSI signal strength", "time difference of arrival" (TDOA) or "angle of arrival" (AoA) are used, although other approaches are also possible.
As in order to have a location accuracy of 2m it is necessary to use multiple receivers (e.g. every 10m) and in addition the active tags usually have a high cost (>100€), this type of active solutions are implemented in very controlled scenarios and for the management of high value assets (e.g. Data Centers).
Also, during these years and for the case of high precision (0.5m), solutions based on Ultra Wide Band (UWB) technologies have been developed. These systems have to our understanding several problems, because they need multiple antennas (less than 10m between them), direct vision and consume a lot of energy, so the active tags have little autonomy. Around these solutions there are manufacturers with new approaches, such as those who are proposing the use of passive UWB tags, but in any case need even more receiving antennas and in this case, energy emitters as well.
Another promising technology to achieve accuracy of <0.5m and that we are currently studying, is the use of ultrasound. In this way, active tags are obtained with greater autonomy. Its implementation and price is similar to that of UWB systems, but has advantages, such as that with these solutions can also locate smartphones through a small program that emits and receives ultrasound through the micro and speaker terminal. There are even scenarios in which they are unique, such as the location underwater.
We are currently validating the possibilities of this new technology. From our pilot tests we see very promising for the projects of Industry 4.0.
However, while these high-precision localization markets are maturing, our company has now opted for the localization of things based on bluetooth tags (BLE). With this type of technology you can achieve the advantages of active RFID, but with fewer antennas and very economical tags. In addition, by using geo-referencing systems based on Bluetooth we can also use these infrastructures with smartphones, achieving a very flexible integration of solutions for locating people and things.
These bluetooth technologies are based on the developments made in recent years around smartphones and have led to an industry of "beacons". Multiple device manufacturers have been launched to the market following the protocols proposed by Apple (iBeacon) and Google (Eddystone).
BLE 4.0 has now become one of the key parts for IoT (Internet of Things), thanks to its omnipresence (smartphones), low power consumption and protocol optimization for low-speed transmission.
It is now possible to obtain cost tags of tens of euros (depending on the configurations) and with a battery life of several years.
These devices have initially been used to generate messages to smartphones (e.g. advertising for the distribution market), although they have also been used to a limited extent for asset management.
The main disadvantage of locating things with this solution was that multiple beacons had to be deployed around the stage (e.g. every 5m), which led to continuous problems with operations, as these devices could be easily removed by users and/or they could run out of battery, so the scenario had to be recalibrated.
In order to solve these problems, in recent years, manufacturers of BLE radio receivers/emitters have appeared to avoid these problems by installing these devices in the same WiFi AP. Thus a few manufacturers already have dual equipment (WiFi and BLE).

Array BLE

We have chosen Mist Systems' solution which has a 16 BLE array at the access point.
This manufacturer has implemented the concept of "virtual beacons". This avoids having to install the beacons around the stage. In addition, their solution, having multiple BLE radios, is self-calibrated to respond to changes in the scenarios.
This design allows to obtain precisions from 1m to 3m with Receivers located at distances of 25m and with low cost BLE tags.
In a complementary way, and as we have already commented before, with this type of solutions we have the "bonus" that we can interact with the BLE of the smartphones of the stage (e.g. via messages) and that we can also locate them.

NB IoT Sensor

Finally indicate that this solution is in line with the industry and will allow the use of BLE/NB IoT tags for the location of assets in interiors and exteriors of buildings. These new tags have an integrated GPS. This way in interiors we will be able to follow them via BLE and in exteriors via NB (Narrow Band) 4G

With the appearance and the push that the NB 4G is going to have, we have discarded working, initially, with other LPWAN systems (e.g. LoRa). We are agnostic about this and if in the future the industry develops dual beacons for other LPWANs we will consider integrating them.

Description of the Aplicaziones solution:

As indicated above, our commercial proposal is based on two architectures that are totally independent, although they can be integrated at the level of the client's operating system:

  • Low-cost people location solution developed by Aplicaziones (R&D project "Veloz") and based on the use of inertial navigation in the smartphone.
  • Solution for locating things and people based on the use of BLE devices. In this case we based the solution of BLE beacons of Mist Systems and third companies that develop sensorized BLE tags.

From our experience, more and more, clients who have needs to locate people indoors also demand the location of things and viceversa. In addition, with the evolution of "Smart Cities" and intelligent buildings towards IoT, an integrated location and sensing solution is becoming more and more necessary.
With the two solutions described in this document, we are in a position to attack these future markets.

Inertial Navigation: Project Veloz

Fingerprint

It is a platform developed by Aplicaziones for indoor location based on the use of smartphone sensors to determine the location of people.

You do not need to install reference beacons (although you can use them if they are available). It is based on a strategy that we call fingeprint and that consists of capturing the information of the possible roads/routes of the scenario by the smartphone sensors. This information is then analysed with the data obtained by the smartphone sensors of the person to be located. The result is compared with the estimation of the inertial navigation algorithms and with the GPS information (if any). Then, by means of a fusion engine, the best estimation of the position is obtained.

In order to achieve a reasonable autonomy of the smartphone, calculations are made in the cloud.

Advantages of the solution:

  • SmartphoneIt's low cost, all you need is a smartphone. All calculation and location services are done in the cloud.
  • As it is its own development, it can be integrated and adapted to customer needs.

Complementarily, in the following links you can see some of the European projects in which we have been involved and how this technology may be applied in the future for several vertical markets:

• Private Security Sector: Civil Protection Security Awareness (CPSA)

• Maritime sector: Crew Situation Awareness (CSA)

• Health sector: Autonomous system for Active Life (AAL) - Positioning of older people by 4 systems (GNSS, RSSI, Inertial, Media)

IoT Location: Locate Project

It is our proposal for the control and monitoring of physical assets / people via Bluetooth.
Each asset/person is associated with a low energy bluetooth tag (BLE).
The location is determined based on the identification of the assets (BLE tags) by beacons that are normally deployed around the area to be controlled.

.

Locate1

In our case we use the BLE beacon solution from Mist Systems. By means of the installation of some systems receivers of bluetooth it allows the creation of "virtual beacons", for the configuration to the measurement of the scenario to control. This avoids the problems arising from the installation of multiple beacons in buildings (calibrations, batteries, etc.).
As indicated above, these BLE receivers are composed of an array of 16 BLE devices in the antenna, allowing greater accuracy in determining the position. There are receivers only BLE and other WiFi+BLE.

Competitive advantages

• It is not necessary to deploy "beacons" in the buildings. Beacons can be virtual (patented))
• By using the array of 16*BLEs per antenna precision is achieved in the location of the orde 1m to 3m without the need for high density antennas
• Can be combined with WiFi installations
•It is a cheaper solution than the current RFID as active tags can be much cheaper and also require fewer antennas.
• Allows the simultaneous location and control of "tags" / smartphones. Therefore, it is possible to locate assets and people.
• Makes it easy to send messages via BLE to smartphones without having to install a SW.
• Allows the reception of information from IoT sensors on stage (temperature, light, movement, etc.) by the control platform.

BLE devices are automatically detected (depending on their refresh period). Smartphones can also be detected via BLE, so they can send near by messages, although it is also possible to use an application on the terminal to improve control and location determination.
It is important to note that it is also possible to detect other types of BLEs devices (e.g. bracelets, watches, tablets, PCs etc.). Here we can see some examples of IoT devices to manage:

:

Sensores

In this way it is possible to develop applications for the control of assets/persons tailored to the needs of each project.

Let's now see a demonstration of our system working simultaneously to locate people with smartphones and IoT devices

 

 

Finally to indicate that to facilitate the commercialization of our technologies of localization in the market, we are realizing alliances with different integrators or manufacturers of systems of:

  • Access Control
  • Presence checks
  • Evacuation systems

Integration with third parties is based on our cloud-based microservice architecture and standard protocols (REST API)

MicroserviciosControl Accesos