This project focuses on developing and implementing algorithms and models for Autonomous Orchard Navigation. The goal is to enable a rover to accurately detect and avoid obstacles while maintaining its path within an orchard. The following tasks are being worked on:
- Utilizing 3D stereo imaging and LiDAR scanning to gather comprehensive data for obstacle detection and path planning.
- Developing advanced algorithms to fuse the data from different sensors and improve the accuracy of obstacle detection.
- Implementing machine learning techniques to enhance the robot's ability to navigate safely and efficiently in dynamic farm environments.
- Annotating large datasets and training the machine learning models to fine-tune their performance.
- Ensuring the robot's robustness and adaptability to different orchard conditions.
- Feel free to explore the code and contribute to this project. Together, we can revolutionize farming with AI!
Thi project aims to address the limitations of current methods for characterizing media bias. The existing approaches are often subjective and lack the level of detail required for an accurate assessment. This project intends to create a purely empirically driven model that characterizes media sources in a probabilistic way, eliminating the need for human subjective judgments. The goal is to provide a comprehensive tool that can assess media bias in a more nuanced and context-aware manner. The following tasks are being worked on:
- Develop a model that probabilistically characterizes media bias using empirical data and advanced research techniques.
- Avoid biases inherent in current bias assignment methods by removing the reliance on human judgments.
- Incorporate sentiment analysis and fact selection as primary drivers for determining bias.
- Utilize natural language processing, topic modeling, sentiment analysis, and data management tools to build the model.
- Create a media landscape assessment tool that can be used to combat information warfare.
- Ensure the model is flexible enough to adapt to different media landscapes, both within the United States and globally.
The objective of this project is to enhance an open-source project called COSMOS to cater specifically to our chemistry research papers. The focus is on autonomous ingestion of PDF files and extraction of relevant information, which will be stored in an online database. The key tasks include:
- Collaborating with the COSMOS community to integrate the necessary improvements for handling chemistry research papers.
- Utilizing Docker to create an isolated environment for running the ingestion process.
- Implementing Dask to distribute the processing load across multiple machines, ensuring efficient ingestion of a large number of PDFs.
- Utilizing XGBoost, a powerful machine learning library, to extract valuable information from the research papers.
- Storing the extracted information in an online database for further analysis and retrieval.
- Feel free to contribute to this project by suggesting improvements, reporting issues, or submitting pull requests. Your input is highly appreciated!
If you're interested in any of the projects mentioned above or have any questions, suggestions, or ideas, I would love to hear from you. Contributions are always welcome! You can reach out to me via email at [email protected].
Thank you for your support and interest in my projects!