Dingqi(Daisy) Zhang
I am a graduate student at UC Berkeley advised by Prof. Mark W. Mueller and
Prof. Jitendra Malik. I am broadly interested
in Robotics, with a focus on extreme adaptative control for aerial systems in challenging
environment with real-world validation. I completed my undergraduate study at Cornell University,
with a double major in Computer Science and Mechanical Engineering.
Email  / 
Resume  / 
Google Scholar
|
|
|
Learning a Single Near-hover Position Controller for Vastly Different Quadcopters
Dingqi Zhang, Antonio Loquercio, Xiangyu Wu, Ashish Kumar, Jitendra Malik, Mark W.
Mueller
ICRA 2023
webpage |
abstract |
bibtex |
arXiv |
video
This paper proposes an adaptive near-hover position controller
for quadcopters, which can be deployed to quadcopters of very different mass, size and motor
constants, and also shows rapid adaptation to unknown disturbances during runtime. The core
algorithmic idea is to learn a single policy that can adapt online at test time not only to the
disturbances applied to the drone, but also to the robot dynamics and hardware in the same
framework. We achieve this by training a neural network to estimate a latent representation of
the robot and environment parameters, which is used to condition the behaviour of the
controller, also represented as a neural network. We train both networks exclusively in
simulation with the goal of flying the quadcopters to goal positions and avoiding crashes to the
ground. We directly deploy the same controller trained in the simulation without any
modifications on two quadcopters in the real world with differences in mass, size, motors, and
propellers with mass differing by 4.5 times. In addition, we show rapid adaptation to sudden and
large disturbances up to one-third of the mass of the quadcopters. We perform an extensive
evaluation in both simulation and the physical world, where we outperform a state-of-the-art
learning-based adaptive controller and a traditional PID controller specifically tuned to each
platform individually.
@article{zhang2023learning,
title={Learning a Single Near-hover Position Controller
for Vastly Different Quadcopters},
author={Zhang, Dingqi and Loquercio, Antonio and
Wu, Xiangyu and Kumar, Ashish and
Malik, Jitendra and Mueller, Mark W},
journal={arXiv preprint arXiv:2209.09232},
year={2022}
}
|
|
Tennibot: An Antonomous Tennis Ball Collector
Dingqi Zhang*, Jerry Tang*
(SP24 Ignite Grant, 3% | FA23 Spark Grant, 9%)
Ball collection has always been a laborous and time-consuming work
in tennis training. We design Tennibot to automate this process, so that players and coaches can
dedicate their efforts entirely to skill development. Tennibot is a mobile robot that can collect
tennis balls autonomously. It is equipped with a roller to collect balls and a cassie platform to
move freely. We have built a prototype and demonstrated its functionality. We are currently
working on adding vision into the system so that the Tennibot can compute optimal paths for
maximum collection efficiency based on vision inputs.
Our proposal has been selected for Ignite Grant in Spring 2024 and Spark Grant in Fall 2023 by the
Jacobs Institude Innovation Catalysts.
|
Personal
In my spare time, I play tennis and
climb. I'm also raising a
cat!
|
|