Welcome to the research page of the Automatic Control Group (RAT) 

Here you will find the core research directions and projects of our group member, who seek to keep up with the rapid pace of developments in areas such as artificial intelligence, machine learning, computer vision, cloud systems, cyber-physical systems as well as availability of low-cost processors and various sensors. With these developments, robotics and autonomous systems have been a game changer in academia and industry. Several companies invest billions of dollars in robotics, and the society has already started discussions on how they can adapt themselves to the revolutionary research activities in robotics and artificial intelligence.
Our ultimate goal: Smarter robots
Our primary research interests and contributions lie within the areas of control systems, reinforcement learning, computational intelligence and robotic vision with applications in guidance, control and automation of unmanned ground and aerial vehicles. In today’s world, it is not sufficient to design autonomous systems that are able to repeat a given task in repetitive manner.  We must leverage state-of-the-art autonomy level towards smarter systems that will learn and interact with their environment, collaborate with people and other systems, plan their future actions and execute multiple time-varying tasks accurately. 
Spatial Perception and Embodied Autonomy Research (SPEAR)

Using evolutionary algorithms and deep reinforcement learning to breed unconventional aircraft "bodies" and "brains," SPEAR aims to demonstrate the benefits of creating unconventional designs that include rotary wing components along with soft and rigid elements. 
Safe, Healthy and Environmental Ship Recycling (SHEREC)

The project aims at revolutionizing safety and environmental standards in the ship recycling industry. The enhancement of occupational health and safety conditions, along with the prevention of environmental contamination by hazardous materials, are recognized as the central goals of this initiative.
RL for Multi-Task Control

Autonomous systems are characterized by their ability to act without human control. Model-free reinforcement learning (RL) algorithms have become popular for controlling autonomous systems due to their ability to succeed without prior knowledge of system dynamics We are interested in the design, analysis, and implementation of novel RL algorithms that enable autonomous systems for long, multi-tasked missions.