Our sincere thanks and appreciation go to the following sponsors, whose support makes our innovative research and exploration possible.
Estimation and Control for Complex Dynamic Systems
Dynamic systems, though often hidden from people's view, are everywhere in the industry and society, critically underpinning technologies including robotics, electric motors, aircraft, vehicles, grid and building structures. A dynamic system may demonstrate complex behavior, making it difficult to understand them or control them for automatic running. The research at ISSL is set to extract necessary information from dynamic systems and leverage the information to tame the complexity. A kind of complexity of particular interest is the uncertainties resulting from unmodeled dynamics, external disturbances, data attacks from cyber space, inherent variability of dynamic processes and sensor noises. Intangible and beyond direct observation, uncertainties can lead to incorrectness or failures in system monitoring, analysis and control. At ISSL, interesting efforts are underway to identify and determine uncertainty through system data analysis that integrates dynamic modeling, estimation and probabilistic machine learning. It is hopeful that the research work can make uncertainties explicitly present before one's view and yield novel control design methodologies of uncertainty-plagued systems.
Advanced Battery System Management
Rechargeable batteries are ubiquitous in contemporary society. They are used at a scale of billions in consumer electronics devices, emergency power supply systems, and auxiliary power units in aircraft, submarines, satellites and space shuttles. They are also widely recognized today to be a critical enabler for moving the world forward into a sustainable energy era, with increasing penetration into the sectors of electrified transportation, renewable energy, and grid energy storage. Regardless of the applications, battery systems, especially those based on lithium-ion batteries, require delicate management to ensure performance, safety and life. ISSL has engaged in addressing key challenges in battery management, including battery modeling, state-of-charge/state-of-health estimation, optimal charging control and energy balancing, with a focus on accurate state monitoring, life prediction and degradation-minimizing operation. The research is backed by an advanced PEC SBT4050 battery tester and supported by continual collaboration with the industry. It has led to a few research findings summarized in journal/conference articles and patents.
Distributed, Collaborative Multi-Agent Systems
Multi-agent systems include a group of independent agents, each equipped with information sensing, computing and communication capabilities, seeking to accomplish an objective or task autonomously and collectively. Through cooperation between agents, they have significant advantages over conventional single, centralized systems, e.g., performing tasks at a level of scale and complexity that is difficult or unachievable for individual agents and responding fast to task and environment changes. These advantages have driven the ever-increasing application of MASs across scientific, industrial, and military sectors, with examples including unmanned aerial, marine and ground vehicle teams, mobile sensor networks and connected transportation systems. ISSL researchers are interested in addressing a fundamental challenge limiting the performance MASs, i.e., distributed information availability. A basic perspective is that more efficient information extraction and spread can be enabled among the agents if data analysis is combined with system modeling. The current research is expected to furnish new distributed MAS control methods for enhancing between-agent cooperation.
PEC SBT4050 Battery Module Tester
Clearpath Robotics Jackal
Quanser QBot 2