Hanyu Jin

Hanyu Jin

MSME-R @ CMU

Posts by Collection

portfolio

UAV Autonomous Navigation in Dynamic Environments

In recent years, lightweight autonomous unmanned aerial vehicles (UAVs) have shown great potential for inspection and mapping in indoor construction sites. However, navigating in indoor environments is complex because of the presence of static and dynamic objects. In this project, we are introducing a navigation framework that enables safe operation in dynamic environments. This report introduces a planning based collision avoidance methodology to generate collision-free trajectories for UAVs using sensor input including camera image and localization. The overall framework aims at tracking a reference trajectory and preventing collision to obstacles. In this report, the methodology for the planner module is introduced and the performance of the framework is demonstrated.

UAV tunnel inspection

Tunnel inspection is critical for maintaining infrastructure safety and reliability but poses significant challenges. Traditional methods expose human workers to hazardous conditions and are time-consuming. This project leverages unmanned aerial vehicles (UAVs) to autonomously inspect tunnels, offering a safer, more efficient alternative.

Characterisation and control platform for pneumatically driven soft robots

Soft robotics is an emerging field with significant potential in applications such as healthcare, industrial automation, and rehabilitation. Pneumatically driven soft robots offer high flexibility, light-weight designs, and variable stiffness capabilities, but their development requires accurate characterization and validation of kinematics, dynamics, and control strategies. To address this need, we developed a comprehensive characterization and control platform for soft robotic systems.

Soft Robots with Densely Reinforced Chambers: A Design, Modelling and Evaluation Framework

Minimally invasive surgeries (MIS) rely on precise and adaptable tools to navigate the intricate confines of human anatomy. Soft manipulators, with their flexibility and bio-compatibility, are ideal for such tasks. However, their non-linear material properties and complex kinematics pose significant challenges to accurate modeling and control. This project focused on developing an analytical model for the kinematics of soft manipulators, incorporating non-linear material properties and formulating an optimization framework. The goal was to enhance their functionality while ensuring safety and efficiency in surgical applications.

Battle City Demo

This project focused on creating a dynamic and engaging two-player game demo inspired by the classic Battle City game. Using C++ and OpenGL, we developed a game where players control two tanks to battle against each other in a simulated environment. The project emphasized realistic physical motion and interactions, including tank movement, bullet trajectories, and collision dynamics, to deliver an immersive gameplay experience.

publications

Characterization and Control Platform for Pneumatically Driven Soft Robots: Design and Applications

Published in IEEE International Conference on Soft Robotics, 2023

This paper discusses the design and control of pneumatically driven soft robots and their applications.

Recommended citation: J. Shi, W. Gaozhang, H. Jin, G. Shi, and H. A. Wurdemann. (2023). "Characterization and Control Platform for Pneumatically Driven Soft Robots: Design and Applications." IEEE International Conference on Soft Robotics.
Download Paper

NavRL: Learning Safe Flight in Dynamic Environments

Published in IEEE Robotics and Automation Letters (RA-L), 2024

This paper introduces NavRL, a reinforcement learning framework for safe UAV flight in dynamic environments.

Recommended citation: Z. Xu, X. Han, H. Shen, H. Jin, and K. Shimada. (2024). "NavRL: Learning Safe Flight in Dynamic Environments." IEEE Robotics and Automation Letters.
Download Paper

Intent Prediction-Driven Model Predictive Control for UAV Planning and Navigation in Dynamic Environments

Published in IEEE Robotics and Automation Letters (RA-L), 2024

This paper proposes a model predictive control framework for UAV trajectory planning and obstacle avoidance in dynamic environments.

Recommended citation: Z. Xu, H. Jin, X. Han, H. Shen, and K. Shimada. (2024). "Intent Prediction-Driven Model Predictive Control for UAV Planning and Navigation in Dynamic Environments." IEEE Robotics and Automation Letters.
Download Paper | Download Slides

talks

teaching