Description
Cooperative Simultaneous Localization and Mapping (SLAM) algorithms have garnered attention in the field of robotics, particularly for their application in challenging and uncertain environments such as those encountered in rescue operations. The use of Software in the Loop (SiL) simulation presents a promising avenue to expedite the development and testing of these algorithms in multi-robot systems, offering a safe and cost-effective solution. A critical aspect of this research is the assumption of unpredictable environmental conditions, which significantly impacts robot coordination and data accuracy [1].
This thesis aims to investigate the effects of these uncertain environments on the performance of cooperative SLAM algorithms in multi-robot systems using advanced SiL simulations. These simulations will incorporate realistic environmental dynamics, sensor modeling, and robot hardware virtualization. The ultimate objective is to derive insights from these simulations to enhance the robustness and reliability of cooperative SLAM algorithms, ensuring effective performance even in the most challenging rescue scenarios.
Requirements
- Strong proficiency in object-oriented programming with C++
- Prior experience with Robotics Operating System (ROS)
Tasks
- Review the state of the art in cooperative SLAM algorithms
- Implement a SLAM algorithm of choice
- Create a SiL simulation model of mobile robots in an uncertain environment, including robot dynamics, virtual hardware and computer networks
- Identify quantitative performance metrics for the cooperative SLAM algorithms
- Evaluate different communication strategies under adverse network conditions using the SiL simulation
[1] Pierre-Yves Lajoie, Benjamin Ramtoula, Fang Wu, Giovanni Beltrame (2021). Towards Collaborative Simultaneous Localization and Mapping: a Survey of the Current Research Landscape. In Journal of Field Robotics 2022.
Supervisor: Hermans
