41530 research outputs found
Sort by
United Methodist Children\u27s Center Playground Improvements
Get PDFThe project consists of multiple pieces that were done as a team to improve the outdoor areas of the United Methodist Children’s Center. The scope of the work consists of placing gorilla hair mulch, making a bench for the kids, removing, grading, and placing artificial turf, and replacing the canvas of a shade structure. There are a few other smaller pieces that will be completed as well but those are the main pieces of the project. The project was executed by teaming up with multiple companies and getting teams of students together to construct the pieces of the project in the timeline we were given. Being a children’s center, there was a challenge with working around their schedule and not interfering with the kids. This project began in February of 2023 and was completed on June 4, 2023. The children’s center was left with a cleaner, safer playground for their students to enjoy during their recess, and the members of the project team left with a new understanding of what goes into placing turf, mulch, building benches, and replacing a canvas shade structure
Deep Semantic Hashing for Aerial Livestock Detection
Get PDFThe goal of this project is to be able to accurately detect and count livestock in footage captured by a drone in real time. The main problems with this arise from the fact that a drone can only carry limited computing resources, and hashing is conventionally thought of as a great method of doing image classification very quickly and thus even on low-power devices. In this project, we use both a Faster-RCNN, which is a state-of-the art object detection model as a benchmark to develop a hashing model that can perform a similar task much more quickly. These two models provide a trade-off between accuracy and speed, where the Faster-RCNN is more accurate and gives precise locations of the livestock in the image, while the hashing is significantly faster but is less accurate and only provides the number of livestock in the image. Given that the dataset is very limited in quantity, we also build a generative network to create more images for the model to train on so that it has a more diverse set of hash codes to reference
Building a Benchmark for Industrial IoT Application
Get PDFIn this project, we have developed a rather robust means of processing and displaying large sums of IoT data using several cutting-edge, industry-standard technologies. Our data pipeline integrates physical sensors that send various environmental data like temperature, humidity, and pressure. Once created, the data is then collected at an MQTT broker, streamed through a Kafka cluster, processed within a Spark Cluster, and stored in a Cassandra database.
In order to test the rigidity of the pipeline, we also created virtual sensors. This allowed us to send an immense amount of data, which wasn’t necessarily feasible with just the physical sensors. The web interface allows users to create as many of these virtual sensors as testing requires.
Once the data goes through the pipeline, it is made viewable on the same web interface. Users can search for key sensors, look through important data, and analyze as necessary.
Our IoT pipeline enables seamless data flow and near real-time analytics. Using industry-standard technologies allows for scalability and reliability, making it suitable for all sorts of data-intensive applications
Exploring Cognition and Affect during Human-Cobot Interaction
Get PDFCollaborative robots (Cobots) have recently gained popularity due to their capability to work collaboratively with human operators. This collaborative relationship has been named under the robotics discipline of Human-Robot Collaboration (HRC), in which humans and robots work together to accomplish a common task while also being in the same physical space. An important part of collaboration is the human\u27s decision-making, which is largely affected by their affective and cognitive state. A cobot lacks this fundamental understanding of the human operator. In this research, we utilize a server-client program to communicate the affective states of a human user to a Raspberry Pi installed within the cobot. Furthermore, this research centers on the stress and focus levels of a user to further enhance Human-Robot collaboration within a manufacturing context
Play Area Utilization Optimization for Room-scale Exploration of Virtual Worlds
Get PDFVirtual Reality (VR) opens up new possibilities for developers to create immersive worlds and experiences. While it’s possible to craft unique and engaging interactive environments with unprecedented realism, the virtual world is constrained by the real one. Current approaches to player navigation in VR applications include joystick controls, teleportation, and motion-based movement. While these methods are effective in certain scenarios to overcome real-world limitations, my research introduces a novel approach that leverages room scale-based movement, with portals, to traverse a given VR world. This work presents algorithms that accurately predict the percentage of play area utilized, and rules to implement typical game elements to allow large scale virtual immersion under real world constraints
