Date: Tuesday, November 4th, 2025

Time: 3PM to 4PM ET

Location: Howey building W 401 or Zoom Link

Juntao He

Robotics Ph.D. Student

Woodruff School of Mechanical Engineering

Georgia Institute of Technology

Committee:

Dr. Daniel I. Goldman (advisor) – Schoold of Physics, Georgia Institute of Technology

Dr. Sehoon Ha – School of Interactive Computing, Georgia Institute of Technology

Dr.  Zeb Rocklin –  Schoold of Physics, Georgia Institute of Technology

Dr. Tony Chen – Woodruff School of Mechanical Engineering, Georgia Institute of Technology

Dr. Baxi Chong  – Department of Mechanical Engineering, The Pennsylvania State University

Abstract:

Elongated multi-legged robots can maneuver through confined and irregular environments that challenge traditional wheeled or few-legged machines. However, their many-segmented, articulated bodies create unique motion planning challenges. This dissertation proposes a 2.5D motion planning framework that enables robust and efficient navigation in unstructured and confined spaces.

The framework integrates geometric and probabilistic modeling of body–environment interactions with reinforcement learning–based controllers that coordinate adaptive body waves for forward motion, turning, and climbing. High-level path planning is guided by depth camera perception, while tactile feedback from antennae and foot sensors enables local trajectory adjustment and collision-aware movement.

The research combines simulation, laboratory experiments, and field validation in pipelines, collapsed structures, and natural terrains. The results aim to advance motion planning theory and practice for high-degree-of-freedom robotic systems, broadening the capabilities of multi-legged robots in inspection, exploration, and search-and-rescue missions.