Indoor positioning for autonomous hospital bed

• Main Author: Koh, Yu Xiang
• Other Authors: Li King Ho Holden
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Mechanical engineering::Mechatronics
• Online Access: https://hdl.handle.net/10356/140245
• Institution: Nanyang Technological University

This FYP project explores the possibility of using BLE for localisation of autonomous hospital beds (AHB). An indoor positioning system (IPS) was developed based on BLE Fingerprint method and experiments were conducted to find out the best suited configuration for the application. An optimal balance between cost and accuracy is found after rounds of experiments and analysis. The optimal system setup has beacons spaced two meters apart and the Fingerprint Server uses Kalman Filter in conjunction with Fingerprint Scoring Type 3 (a comparison algorithm developed in this project) to match fingerprints and return an estimated location to medical staff. With the optimal system implemented, the system provides a stable reading with accuracy of 2.3m with 99.7% confidence. While the bed is moving at speed of 0.75m/s, there is poor localisation accuracy of 10m. However, medical staff can expect the localisation to have a stabilisation time of 5 seconds—time for location readings to reach ±1m of the actual location and attain the static accuracy of 2.3m with 99.7%.