Indoor navigation multi storey strategies for autonomous mixed swarm robot drone (MSRD)

• Main Author: Kew, Shen Yu
• Other Authors: Li King Ho Holden
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Mechanical engineering
• Online Access: https://hdl.handle.net/10356/167588
• Institution: Nanyang Technological University

This project studies, designs, builds, and demonstrates mixed swarmed robot-drone (MSRD) ingress egress strategies and algorithms for navigating in-between unknown multi-storey spaces with object detection technique and navigation stack. Robot-drone term refers to a formation of grounded based robot and aerial robot in a team. The potential of indoor aerial robots such as drones is especially important for applications within large industrial sites, hangars, warehouses, and production lines. Besides drones, it is also possible to have grounded robots in a team to overcome different kinds of challenges in travelling in an unknown indoor environment. In this project, there were 10 runs of robot-drone ingress egress test in simulation world and 10 runs of stairs travelling with drone in simulation world to demonstrate the feasibility of the software and algorithm to perform navigation in multi-storey building. After that, the performance of the software and algorithm were evaluated through 30 runs of experiment in real-world ingress egress test in an unknown indoor environment for two different types of grounded based wheel robots each. The robot types included differential drive robot and Ackermann drive robot. The overall result of the real-world ingress egress test shows 71.67% success rate while experiment conducted in simulation world has 80% success rate in ingress egress test and 50% success rate in stairs travelling test. This result shows the proof of concepts of the algorithm and has shown potential to be implemented in various real-life applications.