Skip to Main Content

Basic Search

Skip to Search Results
 
 
 

Left Column

Filters

Right Column

Search Results

Search Results

(Total results 3)

Mini-Tools

 
 

Search Report

  • 1. Sun, Yufeng Removing Human Assistance to Mobile Servicing Robots: Contact-based Manipulation through Deep Reinforcement Learning, from Simulation to Reality

    PhD, University of Cincinnati, 2024, Engineering and Applied Science: Aerospace Engineering

    Wheeled mobile servicing robots have been applied in various indoor environments, such as hospitals, hotels, airports, and office buildings, to provide a wide range of public services including delivery, surveillance, cleaning, disinfection, etc. However, existing service robots lack the capability to autonomously open doors and charge their batteries. Instead, they often require human assistance for door opening and battery charging, limiting their operational areas and hours. This dissertation research introduces a simple portable device that can be attached to an existing mobile base, enabling the robot to autonomously open self-closing doors and charge their batteries using standard wall outlets without human assistance. This assistive device significantly extends the non-stop operational range and autonomous working hours of mobile robots for indoor services. The portable assistive device comprises a 2-degree-of-freedom end-effector equipped with vision and force sensors. The end-effector is actuated by two servo motors, enabling vertical and horizontal movements for tasks such as operating door handles to unlatch doors and inserting a standard 3-pin 120V power plug into a wall outlet for battery charging. A red-green-blue-depth camera is installed on the end-effector to detect and measure the position of target objects, including doors, door handles, and wall outlets. Additionally, a Pi camera is installed at the front of the end-effector to assist in close-range alignment when the vision of the depth camera is occluded. A 3-Axis force sensor is situated at the base of the end-effector to monitor contact forces during task operations. On the side of the device, a retractable sidebar is provided for pulling self-closing doors, equipped with a Pi camera for upward vision and a 3-Axis force sensor at the base. All sensors and motors are controlled by three small single-board computers, including two Raspberry Pi for motor control and force data processing, and (open full item for complete abstract)

    Committee: Ou Ma Ph.D. (Committee Chair); Donghoon Kim Ph.D. (Committee Member); Rajnikant Sharma Ph.D. (Committee Member); Kelly Cohen Ph.D. (Committee Member) Subjects: Engineering
  • 2. King, Samuel Design and Implementation of Door Opening and Battery Charge Device

    MS, University of Cincinnati, 2023, Engineering and Applied Science: Aerospace Engineering

    Autonomous mobile robots have been widely studied and applied in a wide range of applications, ranging from material handling in large warehouses to disinfecting hospital rooms. These robots have been outfitted with systems design to perform all kinds of tasks but still lack fundamental capabilities necessary for effective operation. Currently, mobile robots do not possess the ability to traverse doorways or charge themselves without human intervention or specialized docking stations. These problems have been studied but most solutions require complex 6+ degree of freedom robot arms or a priori knowledge of the door and environment. The primary objective of this work was to design and implement a system that can extend the capabilities of mobile robots to include opening and operating doors and self-charging by means of a standard wall outlet. The secondary objective of this work was to design a system with a minimum degree of freedom to simplify the control of the system. The Door Opening Battery Charge Device (DOBC) was designed, built, and tested in the Intelligent Robotics and Autonomous Systems lab. The device was designed to mount to existing mobile robots and includes a cartesian manipulator, multi-purpose end effector, side bar for interacting with doors, and a sensor suite which allows for the deployment of a force-vision fusion learning model. In addition to the hardware, a full electronic and power distribution system was designed and built which successfully integrates the new device into the existing mobile base system. The system was successfully built and tested with manual joystick control. The device successfully operated a door handle, opened the door, and deployed the sidebar to grasp the door for opening. This device will extend the capabilities of mobile robots and ultimately make them more effective and efficient.

    Committee: Ou Ma Ph.D. (Committee Chair); Janet Jiaxiang Dong Ph.D. (Committee Member); Shaaban Abdallah Ph.D. (Committee Member) Subjects: Aerospace Materials
  • 3. Almshekhs, Rasha Data Modeling to Predict the Performance of Emerson Walk-in Freezer

    Master of Science (M.S.), University of Dayton, 2017, Mechanical Engineering

    The energy consumption of Walk-in freezers are greatly affected by the door opening. Extensive studies have been made to study the effect of the door opening on the refrigerator/ freezer energy consumption. This paper presents the effect of the door openings of walk-in freezer energy consumption. The walk in freezer was tested in Emerson laboratory to determine the responsiveness of its energy consumption, evaporator return air temperature, and inside air temperature to various door opening schedule. The experiment performed to explore the effect of new door opening schedule on energy consumption and either evaporator return air temperature or inside air temperature. More detailed tests were performed under three different control models. The models have been developed to determine the effect on energy consumption and either evaporator return air temperature or inside air temperature. The testing procedure was conducted under the ambient air temperatures of 70 F, and the door opening operation was carried out by one automatic robotic apparatus in an environmental control room. The three models succeed in eliminating original door opening schedule and predicting new door opening schedule, but they failed in predicting the magnitude of energy consumption and evaporator return air temperature strikes due to door opening. A new model was developed to predict the total power consumption and its results show that increasing the number of door openings will increase the energy consumption.

    Committee: Kevin Hallinan (Committee Chair); Andrew Chiasson (Committee Member); Jun Choi Ki (Committee Member) Subjects: Mechanical Engineering