Author(s): Sebastien Fregeau, Ella Hansen, Thomas Munro,Lucca Coelho, Rodrigo Armaza, Efe Sezer, Efe Kaya
Mentor(s): Masoud Malekzadeh
Institution SUU
Simultaneous Localization and Mapping (SLAM) is a key computational technique in robotics that allows a device to build a map of an unknown environment while simultaneously determining its own position within that map. SLAM plays a critical role in enabling robots to navigate autonomously in diverse and dynamic environments. In recent years, improvements in SLAM algorithms have led to better performance, making it a popular choice for a wide range of applications, including autonomous vehicles, drones, and indoor robots. Our research focuses on developing a robust multi-sensor SLAM algorithm tailored for low-cost, versatile robotic platforms using Raspberry Pi and Jetson Nano. These computing devices are chosen for their affordability and flexibility, making them ideal for integrating with various sensors, such as cameras, LiDAR, and IMUs (Inertial Measurement Units). By combining data from multiple sensors, we aim to improve the accuracy and reliability of SLAM, especially in challenging scenarios where a single sensor may not provide enough information. In this project, we are implementing sensor fusion techniques to merge data from different sources, allowing the robot to create a more detailed and accurate map of its surroundings. For instance, visual data from a camera can help identify landmarks, while distance measurements from a LiDAR sensor provide precise information about the robot’s proximity to obstacles. Additionally, IMU data helps estimate the robot's motion and orientation, enhancing the overall stability and robustness of the SLAM algorithm. The main goal of our research is to develop a SLAM algorithm that can be effectively used in real-world applications, such as warehouse automation, indoor navigation, and search-and-rescue missions. By leveraging the processing capabilities of Raspberry Pi and Jetson Nano, our algorithm aims to offer a practical solution that balances computational efficiency with high accuracy. This approach has the potential to lower costs and make advanced navigation technology more accessible for small-scale and hobbyist robotics projects. Through this research, we hope to demonstrate the benefits of multi-sensor integration in SLAM, paving the way for more reliable and efficient autonomous robotic systems. The findings from our project could contribute to the broader field of robotics by providing insights into how different sensors can complement each other, ultimately enhancing the performance of SLAM algorithms in various environments.