Local Navigation System for a Robot Based on a Network of BLE Beacons

Abstract

The report presents a method for constructing a navigation system for a mobile robot in indoor environments based on a network of beacons using Bluetooth Low Energy (BLE) technology. To determine the position of the mobile robot inside the premises, a network of ESP32 microcontrollers, located at predefined points within the space and operating in BLE beacon mode, is utilized. The position of the robot is calculated based on the analysis of the signal strength from these beacons. This strength is determined by a controller installed on the robot through sequential polling of beacons within its visibility range. To enhance the accuracy of the proposed system, a model was developed using experimental data and constructed with a neural network to describe the dependence of the signal strength (RSSI - received signal strength indicator) on the robot's position relative to the beacon. The resulting model is then used in the Levenberg-Marquardt method for optimal estimation of the robot's position. The main advantages of the system include high accuracy, low energy consumption, affordability, and scalability to environments of various configurations.