CS 525Robot Perception

Catalog

Robot Perception is the ability of a robot to perceive, understand, and reason about the surrounding environment. Perception enables autonomy in real-world robots, including self-driving cars, aerial drones, and autonomous boats. Fundamentals of robotics, sensing techniques, 3D representations, state estimation, sensor fusion, Simultaneous Localization and Mapping (SLAM), and learning algorithms. Open problems in robot perception, including object detection and tracking, semantic scene understanding, pose estimation, perception in dynamic environments, open-world perception, perception for manipulation and navigation, transferring simulation trained models to real world applications (sim2real), and analyzing uncertainty in models with probability theory and statistics (uncertainty modeling). The class is a mixture of traditional lectures, in-class activities, and state-of-the-art paper presentations. Includes one semester-long independent project for which students are taught and encouraged to use AI tools to help improve the efficiency of their coding and project analysis. Prerequisite: CS 375 Design & Analysis of Algorithms, MATH 304 Linear Algebra. Recommended: CS 350 Operating Systems – for knowledge of Linux-based operating systems A background in Computer Vision, Machine Learning, and Robotics is a plus, but not required, and the basics will be introduced in context. Familiarity with the Robot Operating Systems (ROS) will be beneficial. Offered at least once a year. 3 Credits.