Research
Goals and Overview
Our mission is to establish a rigorous theoretical foundation for resilient and intelligent trajectory planning and control algorithms tailored to complex, high-dimensional autonomous robotic systems—including humanoid, quadrupedal, and other legged robots. These algorithms are designed for deployment in challenging applications such as disaster response, industrial automation, and cooperative multi-agent systems with distributed control architectures.
Objectives
We pursue two primary objectives: (1) developing algorithms to systematically design robust and intelligent controllers for high-dimensional, complex robotic systems, and (2) translating these control frameworks into practice through experimental validation on highly dynamic robotic platforms in the ARCOL Lab.
Motivations
In recent years, interest has grown rapidly in designing emergency-response robots and developing machines with human- and animal-like capabilities. Legged locomotion is central to this progress, enabling robots to climb stairs, step over gaps, and traverse uneven terrain more effectively than wheeled systems. Research in this area is driven both by the desire to restore mobility for individuals with disabilities and by the need to assist humans in hazardous environments. Yet, legged robots capable of performing at this level do not currently exist, due in part to limitations in feedback control theory. Although mechanical design and sensing technologies are advancing quickly, a critical gap remains in optimization and control theory for real-time path planning and control of these increasingly sophisticated legged machines.