Controls for Assistive Robots

Guangzheng Zang

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2017-79
May 12, 2017

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2017/EECS-2017-79.pdf

The industry of robotics is on the rise; an increasing number of robots are being used in factories and households. Our capstone project – Controls for Assistive Robots – deals with this market demand. We aim to implement and integrate various subsystems of UC Berkeley’s InterACT lab to enhance its infrastructure. My technical contribution is a robot perception system that gives the robot the ability to “see” and interact with the environment. Specifically, we implemented a vision system using Microsoft’s Kinect v2 sensor. We managed to make the Kinect a quantitative sensor ready to use in research. We also integrated it with other subsystems in the lab and build a system so that the robot could detect the pose of certain objects in real world, and plan to grab them in a 3D virtual environment. With the systems we developed, future researchers in the lab will be able to conduct their own research and keep making a meaningful impact to the robotics industry.


BibTeX citation:

@mastersthesis{Zang:EECS-2017-79,
    Author = {Zang, Guangzheng},
    Title = {Controls for Assistive Robots},
    School = {EECS Department, University of California, Berkeley},
    Year = {2017},
    Month = {May},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2017/EECS-2017-79.html},
    Number = {UCB/EECS-2017-79},
    Abstract = {The industry of robotics is on the rise; an increasing number of robots are being used in factories and households. Our capstone project – Controls for Assistive Robots – deals with this market demand. We aim to implement and integrate various subsystems of UC Berkeley’s InterACT lab to enhance its infrastructure. My technical contribution is a robot perception system that gives the robot the ability to “see” and interact with the environment. Specifically, we implemented a vision system using Microsoft’s Kinect v2 sensor. We managed to make the Kinect a quantitative sensor ready to use in research. We also integrated it with other subsystems in the lab and build a system so that the robot could detect the pose of certain objects in real world, and plan to grab them in a 3D virtual environment. With the systems we developed, future researchers in the lab will be able to conduct their own research and keep making a meaningful impact to the robotics industry.}
}

EndNote citation:

%0 Thesis
%A Zang, Guangzheng
%T Controls for Assistive Robots
%I EECS Department, University of California, Berkeley
%D 2017
%8 May 12
%@ UCB/EECS-2017-79
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2017/EECS-2017-79.html
%F Zang:EECS-2017-79