Exploiting Mechanical and Digital Intelligence for Unusual Soft Robot Locomotion in the Wild
Project Description
Can machines move as efficient as animal? Can soft systems uniquely overcome challenging outdoor environments when other systems fail? Inspired by the remarkable agility of soft-bodied creatures moving across sand, rocks, mud, and cluttered terrain, this project explores how such capabilities can be embedded into engineering systems. To this end, we will combine mechanical intelligence (smart body design) with digital intelligence (adaptive control) to create novel locomotion mechanisms and soft robotic designs. Instead of avoiding contacts, these robots will leverage them to simplify control and improve speed, stability, and robustness. Exploiting this principle, and through a synergy of high-fidelity simulation with hands-on prototyping and experiments, we will target unusual but essential soft robotic behaviors for real-world traversal, from sprinting and contact negotiation to even launching from the ground.
Supervisor
ZHANG, Xiaotian
Quota
2
Course type
UROP1100
UROP2100
UROP3100
UROP4100
Applicant's Roles
(1) Work as part of a research team to brainstorm, search, and identify a locomotion behavior of interest.
(2) Develop conceptual designs that can produce these behaviors through body mechanics and terrain interaction.
(3) Run high-fidelity simulations, optimization, and/or learning-based methods to test ideas, verify mechanisms, and evolve performance.
(4) Build prototype hardware and conduct experiments (field testings). Perform data collection and analysis.
(2) Develop conceptual designs that can produce these behaviors through body mechanics and terrain interaction.
(3) Run high-fidelity simulations, optimization, and/or learning-based methods to test ideas, verify mechanisms, and evolve performance.
(4) Build prototype hardware and conduct experiments (field testings). Perform data collection and analysis.
Applicant's Learning Objectives
Through this project, the student will be able to:
(1) Build an overview of the field of terrestrial soft robotics and locomotion.
(2) Gain theoretical understanding and practical experience with soft-body simulation, nonlinear optimization, and reinforcement learning.
(3) Develop hands-on skills in prototyping, experimental testing, and data collection and analysis.
(4) Develop a collaborative spirit through teamwork and communication, while building an independent research mindset for problem identification, troubleshooting, and learning from failures.
(1) Build an overview of the field of terrestrial soft robotics and locomotion.
(2) Gain theoretical understanding and practical experience with soft-body simulation, nonlinear optimization, and reinforcement learning.
(3) Develop hands-on skills in prototyping, experimental testing, and data collection and analysis.
(4) Develop a collaborative spirit through teamwork and communication, while building an independent research mindset for problem identification, troubleshooting, and learning from failures.
Complexity of the project
Moderate