Grant: Alliance Grants

The goal of this project is to develop a robust vision-based intelligent drone system that can fly autonomously in indoor environments. The project will investigate two critical research issues of the problem, including semantic 3D map building and visual SLAM. The proposed solution is only based on a camera and IMU sensor, thus it can be widely applied to most UAV platforms, especially small drones with limited payload and power constraints. Compared with previous methods, this study offers semantic and contextual information about the environments and proposes to build a landmarks dictionary that significantly facilitates the computation of SLAM and obstacle estimation. Thus, it will increase the intelligence of the system and provide a more efficient and reliable solution than previous systems. The results and findings of this project will also enable many other research topics beyond drones, like self-driving and autonomous vehicles. The project will collaborate with SOTI Aerospace with the endeavor to build a research and development team to develop the next generation of aerial technology in advanced drone and robotics research. In addition to the cash contribution, SOTI has committed substantial in-kind contributions to this project. This project directly aligns with SOTI's strategic plan for ground-breaking research and development of drones. The proposed solution is technically novel and will advance the state-of-the-art performance and applications of drones. The success of this research project will produce new intellectual property and establish Canada as a global hub for cutting-edge aerospace innovation, attract diverse talents and train experienced researchers and engineers to improve the labour shortage issue in Canada. The study also has significant impacts on the economic development of Canada, especially in aviation and transportation, and increases Canada's competitive capability in the global economy.