Shaoting Peng

I am a second-year master student majoring in Robotics at GRASP Lab of University of Pennsylvania, co-advised by Prof. Nadia Figueroa and Prof. Ruzena Bajcsy. My research interests currently lie in learning and perception-based robust, efficient and generalizable human-robot interaction (HRI).

Prior graduate study, I received my bachelor degree in Computer Science and Technology from ShanghaiTech University in 2022 advised by Prof. Xuming He. During my undergraduate, I worked as a research intern at OpenDriveLab under the supervision of Prof. Hongyang Li

Email  /  CV  /  linkedin  /  Github

profile photo

Research Projects

Object Permanence Filter for Robust Tracking with Interactive Robots
Shaoting Peng, Margaret X. Wang, Julie A. Shah, Nadia Figueroa
Accepted to ICRA 2024
project page / video / paper

Proposed a set of assumptions and rules to computationally embed ``object permanence'' into a particle filter scheme to form the 6-DoF object permanence filter (OPF) for occlusion-aware robust perception.

On the Feasibility of EEG-based Motor Intention Detection for Real-Time Robot Assistive Control
Ho Jin Choi*, Satyajeet Das*, Shaoting Peng*, Ruzena Bajcsy, Nadia Figueroa
Accepted to ICRA 2024
project page / video / paper

Investigated the feasibility of using electroencephalogram (EEG) to achieve binary motor intention classification for robot assistive control in real-time.

Camera-Lidar Fusion-based Pointcloud Segmentation in Auto-driving
Research project at OpenDriveLab, Shanghai AI Lab

Explored effective methods for pointcloud segmentation based on camera-Lidar fusion. The RGB image from camera was segmented using DeepLabV3+ to get the segmentation score for each pixel, then projected on lidar pointcloud followed by lidar segmentation. An average mIOU (mean Intersection over Union) of 69.8% across 19 classes on KITTI odometry dataset is achieved, demonstrating a 6% improvement over the baseline model, Cylinder3D.
Cross-modal Weakly-supervised Segmentation for Myocardial Infarction
Research project at PLUS Lab, ShanghaiTech University

Explored a weakly-supervised pipeline to register two myocardial image modalities, which are T2 weighted imaging (T2W) with edema regions and phase-sensitive inversion recovery (PSIR) with infarction regions. Participated in RANSAC flow-based registration, nnUNet-based segmentation, and post-processing to reduce the segmentation uncertainty. The recorded Sørensen–Dice coefficient of 0.68 provides evidence of a correlation between myocardial necrosis and edema regions.


University of Pennsylvania

Master of Science in Robotics
September 2022 - Now

  • Working as a graduate researcher at Figueroa Robotics Lab
  • ShanghaiTech University

    Bachelor of Engineering in Computer Science and Technology
    September 2018 - June 2022

  • Worked as a undergraduate research assistant at PLUS lab
  • Worked as a teaching assistant in Prof. Sören Schwertfeger's CS110: Computer Architecture in Spring 2021.
  • Worked as a teaching assistant in Prof. Lifeng Yang's CS110: Computer Architecture in Spring 2021.
  • Magna Cum Laude (top 10%)
  • University of California, Berkeley

    Exchange Student
    July 2019 - August 2019


  • I love photographing, here can find some of them.
  • I like playing acoustic guitar, my favorite guitarists are Kotaro Oshio and Masaaki Kishibe.
  • I like doing sports, especially working out, swimming, playing badminton and tennis.
  • I'm a cat person, and my MBTI is ISTP
  • I believe in the conservation of luck.

  • This template is a modification to Jon Barron's website.