This seminar series is a means for robotics students to share their research in a more casual, conversational setting. All RRiP’s take place at UH243 at 2pm every Friday during the summer. This Link will take you to the schedule, which may change.
UPCOMING
August 11, 2023
Cameron Schloer (NEST Lab)
August 4, 2023
Davis Catherman (NEST Lab)
PAST
July 28, 2023
Antonio Lopez (NEST Lab)
July 21, 2023
Savita Kendre (Robotic Materials Group)
July 14, 2023
Alex Chilusia (COMET Lab)
June 30, 2023
Khai Yi Chin, Ashay Aswale (NEST Lab)
June 23, 2023
Sreejani Chatterjee (MER Lab), Yash Garje (COMET Lab)
At the Robotics Research in Progress Seminar session on June 23 PhD Candidate (OR student) Sreejani Chatterjee shared her research on keypoint based adaptive visual servoing for vision based robot control in configuration space. She discussed challenges of vision based control such as inaccurate robot models, unstructured environments, and soft or deformable surfaces. She used open source framework such as the DREAM framework for keypoint detection which is commonly used for human pose estimation. While this approach was helpful, she instead implemented her own keypoint RCNN for adaptive visual servoing in 3D space.
Yash Garje gave a practice poster pitch for why changes in laser focus can be useful in laser surgery. He will present the pitch and the poster at the 2023 Hamlyn Symposium on Medical Robotics at the Imperial College London.
June 16, 2023
Joshua Bloom (NEST Lab)
At today’s Robotics Research in Progress Seminar PhD Candidate Joshua Bloom shared his research on multi-agent reinforcement learning in the context of environmental non-stationarity in swarm theory. He discussed different combinations of centralized and decentralized training and execution and introduced the approach of aggregated centralized training with decentralized execution. He used the behavior of ants to illustrate the goals of multi-agent cooperation with minimal communication.
June 9, 2023
Abhinav Gandhi (MER Lab)
June 2, 2023
Tess Meier (AIM Lab)
At the Robotics Research in Progress Seminar session on June 6 PhD Candidate Tess Meier shared her research on a wearable hand exoskeleton for individuals with hand impairments resulting from a brain injury. She explained the complex nature of the injuries as well as the physical limitations that make activities of daily living challenging for these individuals. She introduced the HOPE hand as an assistive hand exoskeleton and the PneuHOPE Hand, a version of the hand exoskeleton that can be worn inside an MRI machine. She just started conducting a human study on brain activation during hand exoskeleton using fMRI. In the future she hopes to use this exoskeleton platform to learn about neurorehabilitation
May 26, 2023
James Akl (MER Lab)
From the presenter:
“In this talk, we propose and develop an automation framework to address the challenges, problems, and opportunities of the metal recycling industry. This is achieved by integrating a variety of components and functionalities into a diverse cognitive architecture. The aim is to endow robotic systems with task-specific autonomy against four main problems found in metal scrap cutting and recycling. These are: (1) Cutting path generation, using viewpoint planning and active perception; (2) Autonomous oxy-fuel cutting, using visual feedback for conditioning and control; (3) Cutting task validation, using learning-based inference via neural network models; and (4) Safe structural disassembly, using sequential decision planning. We formalize and discuss the design and evaluation of each of these functionalities.”
May 19, 2023
Shang Gao (Medical Fusion Lab)
From the presenter:
“Image-guided interventions have become increasingly important in clinical practice, enabling minimally invasive procedures with reduced morbidity and improved outcomes. Photoacoustic (PA) imaging is an emerging imaging technology based on laser-generated ultrasound, which depicts the tissue’s optical absorption, offering the mapping of functional information such as neurovascular anatomy and molecular contrast agents. In this work, we introduce a PA-based necrotic region mapping for cardiac radiofrequency (RF) ablation, focusing on visualizing the extent of ablation-induced necrosis and its lesion distribution. This real-time visual feedback on the growth of the ablation necrotic region is expected to reduce recurrence and procedural complications by avoiding incomplete and excessive ablation, respectively. The experiment result also demonstrated the identification of the ablated cardiac tissue on the in-vivo swine heart model.”