Research Overview

Active Research Areas:

  • Autonomous compliant assembly motion in the presence of uncertainty [ LINK ]

It has long been realized that motion compliant to contact can reduce or even overcome uncertainty in robotic assembly and manipulation. However, autonomous and robust robotic assembly in the presence of uncertainty still poses challenges in research and practice.

Current researchers: Amar Saric, Jing Xiao

Current collaborator: General Motors

  • Autonomous and real-time manipulation and inspection with continuum robots [ LINK ]

Continuum manipulators, inspired by invertebrate structures in nature, such as octopus arms and elephant trunks, do not contain rigid links, can deform, and are passively compliant, which make them apt for manipulating a wide range of objects in unstructured and often cluttered environments. There are many interesting research issues about planning and controlling such robots for real-time manipulation or inspection of objects.

Current researchers: Jinglin Li, Zhou Teng, Jing Xiao

Current collaborators: Clemson University, EPRI

  • 3D object detection and pose estimation for robotic operations [ LINK ]

Object detection is to identify if one or more objects exist in a scene and if they do, where they are located. It is a fundamental, but also difficult problem, not only because of the great variation in appearance caused by different light conditions and viewpoints, but also due to occlusions or complicated backgrounds. However, it is a necessary step for many robotic tasks, such as robotic manipulation of objects. 

Current researchers: Zhou Teng, Jing Xiao

  • Digital microfluidic systems

Digital microfluidic systems (DMFS) are an emerging class of lab-on-a-chip systems that manipulate individual droplets of chemicals on a planar array of electrodes. There are many research issues in both device design and algorithm development for scheduling and routing of droplets.

Current researchers: Zhiqiang Ma, Srinivas Akella

  • Haptic simulation for virtual training, prototyping, and teleoperation [ LINK ]

Effective haptic interaction is increasingly found useful in a wide range of potential applications, from virtual training of surgeons, to virtual assembly, virtual prototyping, and telerobotics. Various aspects of haptic interaction have been studied by researchers, of which one major area is haptic rendering: to simulate and render the force or force and moment exerted to or felt by the human operator via a haptic device from physical interaction with a virtual environment, which may include both rigid and deformable objects.

Current researchers: Sibtanu Raha, Jing Xiao

Current collaborators: Beihang University, Carolinas Simulation Center

  • Human activity recognition for nursing care assistance

It is imperative to investigate intelligent sensing and robotic solutions for elder/patient care to enable a few human nurses to provide quality care to a large number of patients. Recognizing patient behaviors automatically will enable nurses to take informed and timely actions in assisting patients.

Current researchers: Junjie Shan, Jianping Fan, Srinivas Akella, Jing Xiao

  • Image/video understanding and multimedia information retrieval

There is an ever-increasing need for effective data mining of multimedia data for a wide range of applications. However, there are many open research issues.

Current researchers: Ning Zhou, Jianping Fan

  • Real-time robot motion in uncertain and unknown environments [ LINK ]

Navigating a robot in practical environments, where obstacles are unknown and move in unpredictable fashion, is a challenging problem, especially for robots of high-degrees of freedom, such as a mobile manipulator. Despite of much literature in robot motion planning, there are still open challenges in sensing-based on-line planning and execution of robot motions in uncertain and unknown environments.

Current researchers: Sterling McLeod, Jing Xiao

  • Stability Analysis of Assembly/Disassembly Motion

During robotic assembly and disassembly motion of a structure or product, it is important to maintain stability in every step. However, existing robot motion planning approaches seldom consider the issue of maintaining stability. There are open research questions to study. Such stability analysis can apply to other applications.

Current researchers: Sourav Rakshit, Srinivas Akella

  • Topic models for tagged text

A large portion of digital text data contains user-annotated tags. This research focuses on integrating the tag information into the topic model.

Current researcher: Zhiqiang Ma, Srinivas Akella

  • Trajectory coordination of multiple robots

For a team of robots working together with intersecting paths, collision avoidance among them can be achieved through coordinating their trajectories so that two or more robots will not be at the same place at the same time. This is an optimization problem.

Current researcher: Srinivas Akella