Agent localization using sensor trilateration

The use of Wireless Sensor Networks (WSN) has a big impact on many modern technologies, including localization for mobile robotics. Sensors mounted on various devices enable the creation of networked wireless links, allowing mutual communication. This presents an opportunity to use WSN technologies in various fields eg. previously mentioned robotics, environmental monitoring, battlefield surveillance, industry process control, financial marketing etc. The information is gathered by nodes that have a specified (readable) localization through means of GPS or other technologies. Each node responds to a set of described events and reports the information further using technologies like Bluetooth, Wi-Fi, cellular and more. For mobile robotics such nodes (in the form of beacons) can be used to calculate and describe the current position of the robot in closed, repetitively arranged interiors. Many facilities like malls and large area stores are already using Bluetooth Low Energy (BLE) beacons for marketing, transport etc. that can be used for this purpose.

The signal sent out by beacons and received by a mobile agent is usually described in literature by the following parameters:

Gathering that information about nearby nodes allows the mobile agent to calculate the distance to each node, which is used to calculate the position of the robot in the environment using trilateration.

There are several works on the subject of trilateration regarding different approaches to the problem. A popular method is to use the RSSI-based trilateration localization technique. It is calculating the position of the agent based on two following equations:

where variables represent the positions for transmitter nodes , and variables represent the radical plane for sphere intersection for both axes. This intersection can be seen in the following image by Oguejiofor et al. (2013):

Trilateration intersection

In [1] the authors aim to quantify the influences of the mobile trajectory. The authors propose the LMAT algorithm, which solves the problem of path planning on a mobile anchor node. The contributions can be summarized as follows:

  1. Selected deterministic trajectory of the anchor node can guarantee that all sensor nodes receive information about localization to perform localization on self. The optimized movement trajectory significantly guarantees enough sensor nodes to be localized and reduces the average localization error.
  2. LMAT algorithm optimizes the travelling trajectory by adjusting the communication radiusdepending on the deployment area in the WSN.
  3. LMAT algorithm saves enery consumption taking energy and travelling length of the anchor node into consideration. The algorithm is compared to the SCAN, DOUBLE SCAN, HILBERT and SPIRAL algorithms.

In [2] the authors tackle the problem of considerable localization errors when performing calculations based on the RSSI value for the position of unknown nodes for wireless sensor networks. The authors present a method for improved RSSI-based localization through uncertain data mapping. The architecture for the LUDM algoritm can be divided into two main components: The Offline Mapping Generation module and the Online Localization module. The Offline Mapping Generation module consists of the RSSI Tuple Sample Generation sub-module and the Mapping Generation sub-module. The RSSI-Location in terms of distribution is presented as RSSI data tuples and interval data. The algorithm module features are described briefly below:

The subject of Trilateration localization for Wireless Sensor Networks was also raised in recent years for mobile agent localization by other works such as: [3], [4], [5] and [6].


Tell me what you think!