Lab Members

Distributed Autonomous System Laboratory

Current Members  
Post Doctoral Scholars
Dr. Hossein Mohamadipanah  
Hossein Mohamadipanah Dr. Mohamadipanah is working as a Postdoctoral Research Associate in the Distributed Autonomous Systems Lab at Oklahoma State University. He received his Ph.D. from the School of Mechanical and Aerospace Engineering at Oklahoma State University in 2014 and he received his M.S. degree from the School of Mechanical Engineering at Sharif University of Technology in 2010. In the fields of Control, he has been working on Neural Network and Hierarchic Kernel methods. His main interests are Intelligent Modeling and Control, Robotics, and Computer Vision.
Ph.D. Students
Harshal Maske  
Hashal Maske Harshal Maske is a Doctoral student at Oklahoma State University with a Masters degree in Mechanical Engineering from Indian Institute of Technology Kharagpur. After completing his masters, he worked for one year at John Deere and for two years at Defense R&D Organization, India. His current research is grounded in the theory of Bayesian nonparametrics, Reinforcement learning, and Stochastic processes to enable decision making and control in real-world, complex problems, specifically autonomous co-robots, such as the Excavator project.
Allan Max Axelrod  

Blurb

My career goal is to enable the use of autonomous systems in dynamic human environments.  A critical challenge to overcome is automated situational awareness in nonstationary environments.  A key aspect from my research which I expect will be instrumental in moving towards automated situational awareness is the use of models on the information injected into a partially-observable time-varying environment.

Recent Developments

This past Summer, I presented an excerpt of my masters thesis on defining a data-driven Fog of War term and on uninformed-to-informed exploration.  The presentation was received by colleagues at Leeds University, Sheffield University, Hertfordshire University, Technische Universitat Darmstadt, Boston University, and Massachusetts Institute of Technology.  

Masters Thesis

Learning to Exploit Time-Varying Heterogeneity in Distributed Sensing using the Information Exposure Rate considers the data-ferrying problem in the context of synthetic and real-world datasets where no a priori information on the datasets is available to the agent. Due to endurance constraints, the data-ferying agent must learn to prioritize reachable sensors based on their expected informatic value, quantified as Kullback-Leibler (KL) divergence. We formulate and apply Bayesian nonparametric Poisson exposure processes (Pep) for modeling the available KL divergence for each sensing location.  We also provide a probabilistic sampling bound on the accuracy of the Pep; this bound is applied in the EIEIO, RAPTOR, and DUCK algorithms, which are shown to outperform baseline methods in syntehic and real-world datasets.

Contact Information

MS Students
Rakshit Allamraju  
   

Rakshit Allamraju is a Masters Student at DASLAB working with Dr.Girish Chowdhary.

His current research includes solving distributed inference for environment monitoring and event prediction using sensor networks and multiagent autonomous decision making. He is currently leading charge in developing DAS-MAGE, which is a cloud based simulation environment for testing multi-agent decision making and control algorithms.

His past research included planning and decision making under uncertainties in partially observable environments, solving Non stationary Markov decision process problems, by transitional estimations of dynamic environment, applying model based reinforcement learning techniques.  

Ali Abdollahi  

Ali Abdollahi is a Master's student in Mechanical & Aerospace Engineering department. He received one of his Bachelor's degrees in Mechanical Engineering and the other one in Biomedical Engineering. Biomechanics and numerical methods for mechanics of material analysis were his main focus in undergraduate.

 

He has worked on model reference adaptive control and model predictive control started by a NASA project. His other research interests consist of Gaussian Process non-Bayesian clustering application to nonstationary dynamical systems and reinforcement learning methods in time-varying Markov decision processes and contested environments.

 

He is currently working on risk-averse optimization and nonstationary model-based reinforcement learning algorithms for risk-averse adaptive decision making.

Sri Theja Vuppala  
  

Sri Theja Vuppala is a Masters Student with  Mechanical and Aerospace Engineering. His research interests deal with Adaptive Control, Swarm intelligence and their applications.

He is currently working on developing Plug-and-AdaptTM STABILIS Autopilot. He is working on integrating STABILIS on a wide array of commericial UAV and the DAS Golf Cart.

He is working on Model Reference Adaptive Control using online Gaussian Processes and it implementation in Fixed Winged UAVs.

His other interests are development of Hardware-in-the-Loop Simulation Environment and validation of various Autonomy based Algorithms in HITL.

Alex Suhren  
  My work focuses on robotic vision systems.  I have implemented vision systems ranging from corner-detection to scale- and angle- invariant optical flow methods.  My most contributions to robotic vision are the product of an ongoing collaboration with Dr. Doug Gaffin at the University of Oklahoma wherein we're focusing on emulating, robustifying and sparsifying the vision-based navigation system of bees.
Milecia Matthews  
Milecia Matthews

I am working on my MS in Mechanical Engineering at Oklahoma State University.

My current research involves intent communication between pedestrians and autonomous vehicles using POMDPs to learn what actions to expect from pedestrians and how they react to requests given by the vehicle. I am also working on developing a fully autonomous robot capable of maneuvering through viscous fluids while collected data about the fluid using POMDPs.

My research interest are: autonomous vehicles, learning algorithms, and the psychology behind human and machine interactions.  

Hunter Young  
             

I am pursuing a Master's of Science in Mechanical and Aerospace Engineering at Oklahoma State University.

My current research involves the cooperative navigation of multi-agent systems in GPS-denied environments. The goal of this research is to allow a multi-agent (2+ robot) system to navigate from point A to point B with the robots stopping as close to the goal as possible. This will hopefully be achieved by using only the robots' heading, speed, and relative distance.

My research interests are the implementation of control theory fundamentals for the use in general robotic applications to real world problems.

Alumni
Jacob Stockton (Currently at Williams Inc.)
A picture of Jacob Stockton Stockton J., Modular Autopilot Design and Development Featuring Bayesian Non-Parametric Adaptive Control, Master's thesis, Oklahoma State University, Stillwater, OK.
Hassan Kingravi (Currently at Pindrop securities)
Jayadeep Pabbisetty  
   Jayadeep Pabbisetty worked in conjunction with the Eglin Air Force Research Laboratory at the University of Florida and Oklahoma State University on the Multi-agent Cooperation in the GPS denied environments.
Pabbisetty, J., IMPROVING NAVIGATION THROUGH COOPERATION AND PATH PLANNING. Master's thesis. Oklahoma State University, Stillwater, OK.
Ben Reish  
 

Benjamin Reish obtained a Master’s degree at Oklahoma State University with a Bachelor’s degree in Mechanical Engineering from Oklahoma Christian University.  He has worked for the Federal Government in the Engineering office of the Sustainment division of F101 and F118 Turbojet Engines at Tinker Air Force Base for seven years.  He predominantly worked engine trending and diagnostics, test cell correlation, and engine performance along with controls and accessories maintenance and sustainment issues and software maintenance coordination.  

His master's work is in theoretical demonstration of advantages of concurrent learning in the presence of uncertainty in the input allocation matrix of a linear system.

Reish, B., CONCURRENT LEARNING IN THE PRESENCE OF UNCERTAIN INPUT ALLOCATION. Master's thesis. Oklahoma State University, Stillwater, OK.

Visitors
David Seiferth  

David Seiferth is a visiting scholar and Master student at the Technische Universität München in Germany.  He is part of the ongoing collaboration between the Distributed Autonomous System Lab at Oklahoma State University and the Institute of Flight System Dynamics in Munich and is doing research in conjunction with his Master Thesis for six months at Oklahoma State University.  During his Master degree in mechatronics and information technology he has worked for six months at Diamond Aircraft Industries in Wiener Neustadt, Austria, where he designed and developed a flight control system and an autopilot for a hardware-in-the-loop simulation of Future Small Aircrafts.  At the Department of Mechanical and Aerospace Engineering his work concentrates on nonlinear control of stochastic systems, especially model predictive control in combination with Gaussian Processes.

-Visiting Stillwater, OK, USA from TU Munich, Germany

Maximilian Mühlegg  
  -Visited Stillwater, OK, USA from TU Munich, Germany
Undergrads
Dane Johnson  
  Leading the DAS-Stabilis Development team
Lucas Kinion  
  DAS-Stabilis Development team
   
Undergrad Alumni
 
   

 

Slideshow Title: 
Labmembers
Slideshow Subtitle: 
Yes the computer is a member.
Slideshow Image: 
Skynet 4.0 (3.0 no good)