Project AI-70: Monitoring and Control System for C-Quester Carbon Capture
Project Client: C-Quester, Inc.
Project Description: The iCapture addresses the critical need for cost-effective and efficient data monitoring and control systems (MCS) in carbon capture operations. It is capable of communicating with various sensor signals, collecting their data, and outputting the sensor data to an online API. Developed in collaboration with Mitico, a company that offers accessible carbon capture technology and services, iCapture seamlessly integrates with their existing sensor networks, providing real-time insights into crucial parameters of their carbon capture system. The iCapture’s automatic sensor configuration, range communication of beyond 100ft, error handling for sensor disconnection or malfunction, and LCD display for displaying error notifications ensure seamless operation and scalability, all while maintaining affordability at under $500. By lowering the cost and complexity of MCS, we enable Mitico to deliver more accessible and sustainable carbon capture solutions, furthering their mission to combat climate change and promote environmental stewardship.
Project CG-23: Remote Sensing for Forest Fires
Project Client: UBC Cloud Innovation Centre
Project Description: FireWatch is an innovative solution engineered for the UBC Cloud Innovation Centre, designed to combat the increasing threat of wildfires in British Columbia, Canada, and across the globe. This product offers early prediction and detection of wildfires, enabling rapid firefighting responses to protect communities at risk from fire outbreaks and associated smoke hazards.
A standout feature of FireWatch is its groundbreaking approach to overcoming connectivity challenges in remote areas. Traditional Wi-Fi and cellular networks often fall short in vast, uninhabited lands prone to wildfires. FireWatch overcomes this by incorporating satellite data and using low-power wide-area network protocols, and ensures reliable wildfire monitoring capabilities even in the most isolated regions. This innovative connectivity solution allows for uninterrupted surveillance and early detection, making it a vital tool for wildfire management.
In addition, FireWatch is designed for scalability and is open-source, making it an economical and adaptable platform for widespread application and future innovation. Its advanced features promise enhanced protection for at-risk communities and encourage collaborative development in wildfire prevention and control.
Project HA-60: Development of Renewable Energy Assessment Tool Mine Site
Project Client: Fluor Canada
Project Description: Purpose of your project: The problem is that the mining industry is a large contributor of greenhouse gases that cause global warming. It hasn’t transitioned to green energy as urgently as other industries. PURE Sim is a desktop and web application that provides a user-friendly experience to retrieve weather data and run simulations of renewable energy systems for feasibility analyses. This project’s client is Fluor Ltd.
Major design contribution: The technical challenge resolved is that engineering planners are better able to assess whether using green energies is feasible compared to burning fossil fuels on a mining site. PURE Sim calculates the total energy output from historical weather data and total installation & maintenance cost depending on user inputs. If a mining site could feasibly incorporate solar or wind energy, then it will help the mining industry better transition to green energy.
Khalil Safar: (604) 700-9495; Puneet Chopra: (236) 877-8384; Ryley McRae: (778) 257-8775; Will Chen (604) 787-7922
Project PN-01: AI Based “Co-pilot” to Assist Solar Project Manager with Crew Assignment and Scheduling
Project Client: Scoop Robotix Inc.
Project Description: Purpose: To create a machine learning model, capable of assisting project managers with assigning crew and staff to various solar installation projects.
Major Design Contributions: Machine learning model built from the ground up by our capstone team: Our team constructed a sophisticated computer program that learns from data to assist in assigning crews and staff to solar projects. Think of it as teaching a computer to recognize patterns and make decisions similar to how a human would.
Utilized scikit-learn Python library for the AI model: We used a powerful toolkit called scikit-learn in the Python programming language to build and train our machine learning model. It’s like having a toolbox filled with specialized tools that help us create smart algorithms.
Back-end hosted on AWS Sagemaker: We deployed the core of our system, where the heavy lifting of computation happens, on a cloud service provided by Amazon Web Services called Sagemaker. It’s akin to renting a supercharged computer in the sky, which allows our system to process large amounts of data efficiently.
Front-end graphical user interface designed to fit specifications laid out by the client: We crafted a user-friendly interface that project managers can interact with easily. It’s like designing the dashboard of a car, ensuring that all the controls are intuitive and accessible, tailored specifically to meet the needs of our client.
Developed REST APIs to integrate the front-end and the back-end: We built a bridge that connects the user interface to the core functionality of our system. This enables smooth interaction between the project managers and the AI-powered scheduling system.
Project SF-51: IoT Solutions for Sustainable Beekeeping
Project Client: Foundation of the Energy Collective
Project Description: Envision a world where the intricate balance of floral brilliance and vibrant landscapes is preserved through innovative technological solutions. Our IoT Solution for sustainable beekeeping leverages a comprehensive array of tools, including a a dynamic website platform, and an arsenal of cutting-edge sensors – including microphones, temperature, weight, and humidity sensors.
Through seamless integration, our solution equips hobbyist beekeepers with real-time insights into the health and well-being of their hives. The sensors, strategically positioned within the hive, continuously collect vital data on environmental conditions and hive activity. This data is cataloged and stored within our database, facilitating comprehensive analysis and trend identification.
Accessible via our website, beekeepers gain instant access to critical hive metrics, enabling proactive management and swift intervention when necessary. Furthermore, our platform fosters a vibrant community where beekeepers can exchange knowledge and share best practices.
Together, armed with technology and expertise, we can forge a future where bees thrive, ecosystems flourish, and the beauty of nature endures.
Project TL-30: Mapping Below the Forest Canopy (Part 1 – Software)
Project Client: Korotu Technology Inc.
Project Description: The health of forests worldwide is fast deteriorating due to increases in deforestation and climate change. In order to make sound planning and management decisions, we must first better understand the present condition of our forests. Tree counts, biodiversity counts, growth stages, and other inventory data all provide extensive information on the state of our environment, yet much of this data is hidden beneath the forest canopy.
Today, traditional forest surveying techniques often involve deploying teams of skilled workers to collect data samples in remote locations, costing substantial time and resources. Expenses are estimated to average around $10,000 per hectare. While satellites and drones may provide information on forest conditions above the canopy, they are currently unable to gather data on the state of trees growing underneath.
To address these challenges, we have partnered with Korotu Technology Inc. to create ForestFolio, a mobile application that aims to reduce the financial and time burdens of data collection, increase the quality and precision of data samples, and make forest surveys more accessible to smaller groups interested in gathering forest data.
ForestFolio combines the various tools needed for traditional surveying into one device by making use of the camera, LiDAR sensor, gyroscope, and accelerometer on a mobile device. The application guides the user through the data collection process for the location, height, diameter, and species of each tree within a fixed-area forest plot..
First, the user is required to walk around the plot while holding up their mobile device to scan their surroundings and generate a map of detected trees. At this step, the diameter of each tree is extracted through machine learning algorithms, eliminating the need for individual tree diameter measurements.
The generated map helps the user keep track of their own position within the plot and the location of each tree. When the user navigates to a tree, they first receive step-by-step guidance on measuring its height. They then input additional data such as the tree species and relevant notes. Once all trees have been completed within the plot, the application waits for a stable Internet connection before uploading the completed plot to a server. There, the data is further analyzed and presented to the user via a website interface, with the option to export the data as a spreadsheet file.
In addition to the mobile application, data collected through other means such as via a drone with a mounted LiDAR sensor can also be analyzed by ForestFolio. This data simply needs to be uploaded to a website interface which will process it and return valuable tree inventory data.
ForestFolio not only eliminates the need for external forest survey tools, but also facilitates the data collection process so that untrained users can quickly begin gathering accurate, high-quality data beneath the forest canopy.
Project TL-31: Mapping Below the Forest Canopy (Part 2 – Hardware)
Project Client: Korotu Technology Inc.
Project Description: The Kanohi T100 is a hardware sensor that can be used to drastically speed up the measuring of forest biomass. It is able to achieve this by using LiDAR technology to create a 3D map of the forest. This device was developed for our client Korotu Technologies to be used by a variety of forestry and conservation groups. During the project, we solved numerous technical challenges including the development of software to interface with the LiDAR sensor, navigation algorithms for keeping track of position within the forest, and custom hardware to interface with the various sensors. We also developed a mechanical packaging and software user interface to improve the user experience. All of these technical developments allowed us to design, integrate and test a fully functional hardware unit able to collect 3D forest data efficiently.
Project TL-63: Measuring Forest Change with AI
Project Client: Korotu Technology Inc.
Project Description: There is currently a market gap in the forest preservation space where land protectors are unable to access affordable forest change and threat analysis tools. Our project addresses this issue by providing a low-cost web app that utilises machine learning to detect forest change from satellite imagery. This project will allow land protectors across Canada to observe and analyse Canada’s vast forested landscape so that the critical services and biodiversity provided by forests can be preserved.