(pending update information)

Project Overview: CropScan focuses on the development, implementation, and management of aerial drones equipped with high-resolution cameras, multispectral sensors, and data analysis algorithms. These drones will be deployed to capture detailed imagery and collect spectral data from crop fields, enabling farmers to identify and assess crop health conditions accurately.

Objectives:

Early Disease Detection: The primary objective of Project CropScan is to enable early detection of crop diseases. Aerial drones will systematically fly over crop fields, capturing high-resolution images and spectral data that can reveal subtle changes in crop health caused by diseases. By identifying diseases at an early stage, farmers can implement targeted interventions and minimize the spread and impact of diseases.

Rapid and Wide-Scale Monitoring: Aerial drones offer a unique advantage by providing rapid and wide-scale monitoring capabilities. They can cover vast areas of farmland efficiently and collect data in a fraction of the time it would take for manual inspections. This allows for the timely assessment of large crop areas, facilitating proactive decision-making and effective disease management.

Spectral Data Analysis: The project will employ advanced spectral data analysis techniques to extract meaningful information about crop health. Multispectral sensors on the drones will capture data beyond the visible spectrum, such as near-infrared and thermal imagery. These data layers will be analyzed to identify specific disease indicators, such as variations in chlorophyll content, canopy temperature, and vegetation indices, providing insights into the presence and severity of crop diseases.

Data Integration and Interpretation: CropScan will develop a robust data integration and interpretation framework to process and analyze the collected aerial imagery and spectral data. Machine learning algorithms and remote sensing techniques will be employed to correlate disease patterns and symptoms with known disease profiles. This will facilitate automated disease diagnosis and provide farmers with actionable information for targeted treatments and preventive measures.

Farmer Advisory System: To enhance the project’s impact, CropScan will develop a farmer advisory system that delivers real-time disease alerts and recommendations. Based on the drone-collected data and disease analysis, the system will provide personalized notifications to farmers, advising them on the most appropriate actions, such as applying specific fungicides, adjusting irrigation practices, or implementing crop rotation strategies.

Farmer Training and Capacity Building: CropScan recognizes the importance of empowering farmers with the necessary knowledge and skills to utilize drone-based disease detection effectively. The project will conduct training programs and workshops to educate farmers on the interpretation of aerial data, disease management strategies, and best practices for integrated pest management.

Conclusion: Project CropScan represents a game-changing approach to crop disease detection, utilizing aerial drones and advanced data analysis techniques. By providing early detection and monitoring capabilities, this project aims to enable farmers to respond promptly to crop diseases, minimizing economic losses and reducing reliance on broad-spectrum pesticides. CropScan paves the way for a more sustainable and precise approach to disease management in agriculture, fostering increased productivity and resilience in the face of crop health challenges.

In pursuit of our mission to study Human Robot Interaction (HRI), this project explores Cognitive Load Theory (CLT) and its impact on a human’s mental processes when operating in complex, high risk, and/or high stress scenarios. The chief impact sought is in studying the limits of cognitive load on a human being and exploring potential avenues for autonomous systems to offset this load.

Through the use of a Virtual Reality (VR) system a comprehensive simulation of operating a lunar rover on the moon was created, complete with low-gravity conditions and a custom seat featuring extensive controls and vibration mechanisms to maximize physical immersion. A subject operating the simulation would navigate the rover on a procedurally-generated lunar surface to a series of random destinations to drop a “marker”, avoiding crashes and arriving within a certain time limit while simultaneously monitoring their oxygen, engine heat, communications channel, and fuel levels. These trials would start with a generous time limit and resources, increasing in difficulty as more destinations are reached successfully.

During each trial the subject’s biosignals such as heart rate and skin temperature are monitored and recorded with the intention of observing their stress response as their challenge and cognitive load increases. 

This project’s results were initially submitted in 2021, presented at the International Conference on Intelligent Robots and Systems (IROS) in 2022 and presented at the NASA Human Research Program Investigator’s Workshop in Galveston, Texas in 2023.

A full PDF copy of the 2021 submission can be downloaded and reviewed here.