Center for Robotics Research Projects
T-shirt Cannon
By Daryn Meadows, Daniel Arguelles, Steven Randall on phase I - Firing mechanism and cannon design
By Joseph Domhoff, Graham Donhowe, David Kuebel on phase II - Drive system and autonomous system design
The purpose of this project is to design and automate a robot with shooting cannon to launch memorabilia at UC fans during athletic events. More specifically, the T-shirt cannon will be used by the University of Cincinnati cheerleading team at Nippert Stadium and 5/3 Arena for football and basketball games, respectively.
The Development of Trench Drain Cleaning Robot
By Adithya Kaushik
The purpose of this project is to develop a novel method for the cleaning of trench drains. It deals with the design, conceptualization, analysis and testing of a robotic device which fits into the drain and cleans it without needing to interfere with the surrounding traffic or the drain itself. The project was an effort to shift the cleaning process for the trench drains from manual to either semi-automated or automated methods.
Due to the previous methods like manual cleaning and waterjet cleaning being slow and highly inefficient processes respectively, there was a need for an automated mechanism which was fast, reliable and did not waste resources. The transition was intended to ensure a safer process with lesser manpower required in addition to being easier to implement. Over the course of developing this robotic cleaning mechanism, two complete prototypes were designed and implemented in the drains. The performance of both the models has been discussed in detail.
Tick Collection Robot for Rough Outdoor Terrain
By Chen Chen, Jorge Benito Montejo, Niki Harrison, Qiu Yesiliang
The objective of this project is to develop a rugged and lightweight robotic platform that acts autonomously, following a course outlined in a convenient graphical interface via wireless user device, to collect ticks in a variety of outdoor terrains, ranging from relatively flat areas of leaf litter to tall grass to heavily forested areas containing numerous obstacles and rough, uneven terrain. The robot will perform active tick sampling (via flagging or dragging procedures) or passive tick trapping.
It will make use of CO2 or other chemical attractants by way of a centralized dispersal system, allowing it to attract ticks to its traps as well as to its flag/drag cloth. The robot will have four distinct types of environmental sensors to facilitate navigation and decision making: one GPS localization unit, one 360-degree laser point scanner or LIDAR system, a set of passive tactile collision sensors, and a set of active tactile sensors.
Development of Robotic Cell for Removing Tabs from Engine Blade
By Prateek Sahay,
The purpose of this project is to design and prototype an automated robotic part process system for completing a few final process steps of machining engine blades. The targeted final steps include reading and checking part identification, marking new information, cutting the tabs, and polishing the cutting areas.
The focus of this project is on cutting off two tabs from the engine blade and make sure that the cutting surfaces are blended well with part edge. The ultimate goals of this automation research are to eliminate human errors and reduce cycle time in the process.
EMG Controlled Soft Robotic Bicep Augmentation
By Jiayue Zhang, Daniel Vanderbilt, Ethan Fitz
Industry workers’ jobs often require them to lift heavy objects. Repeated heavy loading of their arms throughout the workday can lead to muscle or tendon injuries. Hence, the objective of this study is to develop a wearable soft robotic arm enhancement device that will share lifting loads with the user’s muscles to increasing their lifting capacity and reducing fatigue. This project involves developing a soft actuator, a control system, and a method of attaching the actuator to the user’s arms. The McKibben inspired pneumatic muscle acts as the soft actuator of the prototype. EMG control is used for the controlling system. Setups and results of testing for the pneumatic muscle, the controlling algorithm, and the overall performance are discussed. Based on the experiment data, the wearable device prototype can help workers work for a longer time duration, but the improvement of strength varies from person to person.
Using Deep Learning Semantic Segmentation to Identify Lost Seniors
By Pradyumna Elavarthi
The purpose of this project is to design and develop an image segmentation method to identify the seniors lost in a senior community village. The live videos or images are fed from the drones hovering over the village once a lost senior is reported.
A Fully Convolutional Neural Network will be built and trained with the data from Unity Simulator to perform semantic segmentation.
Automated Resistor Element Assembly Cell
By Mark A. Bohman
The resistor element pin spacer assembly (“pinning”) that is currently performed at Post Glover Resistors, Inc. is a physically taxing and time-consuming process and it puts various unwanted physical stress on the assembler.
Two of the parts are sharp-edged stainless-steel plates that expose the worker to cut hazards, and these plates are held apart by a ceramic pin spacer that is held in place by push nuts on the outer faces of the plates. Due to the time-consuming nature of the current process, and with an emphasis on worker safety, Post Glover Resistors, Inc. needs an automated solution for the resister element assembly work.
Two of the parts are sharp-edged stainless-steel plates that expose the worker to cut hazards, and these plates are held apart by a ceramic pin spacer that is held in place by push nuts on the outer faces of the plates. Due to the time-consuming nature of the current process, and with an emphasis on worker safety, Post Glover Resistors, Inc. needs an automated solution for the resister element assembly work.
The automated resistor element assembly cell was developed using a SCARA robot from ABB, three feeder systems with three tracks to deliver pin spacers and push nuts, a pneumatic system to actuate the end effector fixture, and a three gripping mechanism fixture to pick and place the elements, pins, and spacers. The robot programming was done via ABB’s RobotStudio and the FlexPendant.
Design an End Effector for Drive-Through Gas Station
By Khada Gautam
Design a Bionic Keyboard for Erb-Duchenne Palsy Individuals
By Joshua Grubbs, Mark Minutolo, Dillion Hilgeford (Computer Engineering)
Plant Propagation Robotic System
By Noah Kohls, Nick Galbraith, Bryan Wilson
The goal of this project was to develop and integrate a new robotic system to automatically propagate various sub-species of boxwoods. The current process for propagation is completed manually and requires a worker to take several thousand cuttings, organize them, and individually stick them into soil. This is both time consuming and arduous. A SCARA robot was built with its quickness for pick and place applications. The robot was designed so that it could be 3D printed, which allowed for quick prototypes and design iterations. The robot is programmed through an open loop control system and requires occasional resetting to verify the correct location. In order to fully automate this process, a feeder system is built to supply a constant feed of plants to the robot, requiring no manual effort and little wait time. The feeder system works through a process of looping with a refeed on failure. This means that the cuttings will repeatedly go through the system until they arrive to the robot in the correct orientation.