#Automation #Solidworks #Arduino #C++ #3D-Printing #Lathe #Welding
Designed and built an automated system from the ground up to quickly extract soil samples. Manage to reduce sample extraction time from 30 minutes to less than 5 minutes. The device has helped researchers to improve turf condition in 3 golf courses across Nebraska, preserve the environment, and save thousands of dollars for the resort owners.
Refurbished Robotic Arm
#Robotics #Solidworks #Arduino #C++ #3D-Printing
Initially built in the mid-1980s, the Scorbot ER-VII Robotic Arm has a controller with a volume of over 1992 cubic inches and weighs approximately 42 pounds. Moreover, the user interface of the controller is utterly outdated and does not utilize the programming languages that are commonly used today. My team and I were tasked to redesign and build a new controller module using modern-day electrical components and programming languages while maintaining the functionality of the entire system. By using only 30% of the allocated budget, the full functionality of the arm was maintained while the weight and size of the controller were reduced by more than 90%.
Plant Nursery Bot
#Robotics #Solidworks #Arduino #C++ #3D-Printing #LaserCutting
Designed and developed a robotic system for the 2019 ASABE Robotics Design Competition. Participants were challenged to design a robot capable of conducting autonomous inventory management in a plant nursery simulated by a 4 ft. x 16 ft. board. During the competition, teams had to simultaneously identify, collect, and transport ten pots within a five-minute time frame. My team and I were able to secure the 4th position for the advanced division by the end of the challenge. The system is currently being used as a teaching tool for new students.
Cloud Detection & Phenotyping
#MATlab #ComputerVision #MachineLearning
Cloud cover remains a challenging problem for farmers who rely on aerial imagery to maintain the health of their crops. The following project explored ways to detect clouds on these images and how they would affect the accuracy of vegetation indices, such as NDRE, when used to prescribe fertilizers. Various methods in computer vision and machine learning were utilized to accomplish these tasks. As a result, we were able to successfully detect clouds and shadows 92% and 83% of the time, respectively. The goal of this project was to alarm decision-makers about the presence of clouds and shadows in aerial imagery and enable them to account for the effects imposed by them when prescribing fertilizers, thus saving the environment while also producing healthier crops.
#Python #ComputerVision #DeepLearning
This project developed and trained a network from scratch to automatically classify animal species from the Snapshot Serengeti camera-trap image data set. More specifically, this study aimed to alleviate the burdens imposed by non-ideal images when labeling the SS data set by automatically detecting wild animals from motion-triggered camera-trap images using deep learning methods. Several classification experiments were conducted to evaluate the performance of the model. As a result, it was observed that the model yielded high accuracy during training but performed poorly on the validation and testing sets. This project was part of ongoing research at the University of Nebraska, Lincoln.
#Robotics #Solidworks #Arduino #RaspberryPi #C++ #3D Printing
The goal of this project was to build a robotic system for the 2018 ASABE Robotic Design Competition from the ground up. The participants were tasked to design a system to simulate harvesting apples (represented by ping pong balls) on an 8 ft. x 8 ft. playing field. The robots were required to differentiate and harvest eight mature apples (red-colored ping-pong balls), remove eight rotten apples (blue colored ping-pong balls) and leave eight immature apples (green colored ping-pong balls). There were a total of 24 apples on the board. Although this was my first time using computer vision in a robotic project, my team and I were able to secure the 5th position during this competition.
#Robotics #Solidworks #Arduino #C++ #3D Printing
This project was in conjunction with the 2017 ASABE Robotics Design competition. The robotic mechanism was built to simulate a full primocane suppression and selective floricane removal pruning operation. The challenge required teams to develop robots that can identify primocanes and floricanes, selectively cut irregularly distributed quantities a given row density, and remove the derbies from playing field. My team and I were able to secure the 2nd position at the end of this challenge.
Prosthetic Knee Joint
#Solidworks #FEA #3D Printing
The goal of this project was to reach a novel and unique approach in designing a prosthetic knee for above the knee amputees. Using a compliant rolling element (CORE), the two bodies of the upper leg and lower leg are joined to provide an artificial knee replacement that is user-friendly, low-cost, safe from hazards, and lightweight. The CORE prosthetic discussed in this study is able to perform near-natural range of knee motion while during flexion and constrained in extension.