CSCI 682  Topics in AI - Sensor Fusion  3 Units  

Prerequisite: Classified graduate standing.  

Typically Offered: Spring only  

Further study of selected advanced topics in artificial intelligence as presented in recently published journals; possible emphasis on research interests and/or projects of faculty in the department. Consult the Graduate Coordinator to determine how many units may be counted toward your major.

Grade Basis: Graduate Graded  

Repeatability: You may take this course for a maximum of 3 units  

Course Attributes: Graduate Division; Laptop required  

Course Description: This course emphasizes a balance of top-down and bottom-up theoretical understanding of sensor networks, sensor system and information fusion methods, and continuous monitoring of systems and operational environments to provide situational awareness.  Algorithmics and mathematics required to acquire sensor data from concurrent and redundant sensors, both active and passive are studied. This includes review of current research topics on emergent methods and how to use the methods to support higher level application development for applications like self-driving cars, environmental monitoring, and autonomous systems. 

Goals: Students will review research papers, software examples, and case studies to understand fundamental sensor fusion approaches including simple correlation, Bayesian and Dempster-Shafer models, machine learning, and Kalman filtering.  The course will use MATLAB and OpenCV for top-down algorithm and programming as well as research papers focused on fundamental methods of applications of them.  At the end of the course, all students must either draft a research quality survey paper or demonstrate a prototype software that fuses data from two more sensors to improve detection and monitoring systems.

Textbook: Not required, course is taught primarily from notes.