ECE 6970 Bio-Inspired Coordination of Multi-Agent Systems
Lecture: TR 1:25-2:40pm Phillips Hall 213
Lab: T 5:00-7:00 PM Phillips Hall 217
Multi-agent systems exist on all scales in chemistry, molecular biology, entomology, and more recently, robotics. Here, dynamic interactions between large numbers of relatively simple individuals result in complex emergent global patterns. Take as an example the amazing coordination that occurs between thousands of bees in a hive, or millions of cells joining to form hierarchical organ systems with embedded parallel and distributed functions. The success of these systems is typically associated with robustness due to redundancy and the ability to adapt to changing environments, making this approach of great interest to robotics.
This class will survey state of the art approaches to the design of robust behavior in robot collectives. We will focus on coordination and control, agent design, and sources of inspiration from nature. Topics include: swarm intelligence, decentralized vs. centralized systems, distributed consensus, behavior-based robotics and physical intelligence, self-organizing and self-assembling systems, modular robotics, and special topics related to collective transport and construction. The class will be based on both paper discussions and a major research project.
Students taking this course will gain experience reading and reviewing scientific papers, and practice strong presentation skills both oral and written. The class will cover work done by several of the robotics faculty at Cornell University. Finally, students will experience the process of designing and conducting a research project related to robotics, modeling and simulation, or the study of insect collectives in nature.
The class is limited to 15 students, due to the discussion nature of the class and supervision of final projects. Please make sure to come to the first class and fill out the online class form.
Find more infomation here.