Project number: CS34
Project Source: School of Computer Science
Project Title: HomeAgent: A Sandbox for Simulating Smart Home Interaction
and Automation Using Large Language Models
Project Description and Scope: This project involves upgrading the HomeAgent
sandbox framework, designed for simulating smart home environments and
interactions with large language models (LLMs). HomeAgent currently
integrates family agents and a HomeBot to emulate realistic human behaviours
and smart home automation. The goal of this project is to enhance the existing
framework with new features and improvements.
Project Scope:
• Upgrade to GPT-4o:
o Objective: Update the underlying model of HomeAgent to GPT-4o.
o Details: Integrate GPT-4o into the HomeAgent framework,
replacing the current model to leverage the advanced reasoning,
tool use, and human interaction capabilities of GPT-4o.
• Add Family Agent Functionality:
o Objective: Enhance the sandbox environment to support
additional family agents.
o Details: Modify the simulation environment to allow the inclusion
of more family agents, enabling a more comprehensive simulation
of household interactions and dynamics.
• Extend Simulation Timeline:
o Objective: Extend the simulation capabilities to cover a full week.
o Details: Update the simulation timeline to allow scenarios and
interactions to unfold over a week, providing a more detailed and
extended observation of agent behaviours and smart home
automation processes.
• Improve User Interface:
o Objective: Enhance the user interface to display environment data
and agent actions more effectively.
o Details: Redesign the user interface to provide real-time updates
on the smart home environment, including temperature,
humidity, and IoT device statuses, as well as detailed logs of agent
actions and interactions. Expected outcomes/deliverables: Enhanced realism and complexity in smart
home simulations. Improved interaction capabilities with more family agents.
Comprehensive data collection and analysis over extended simulation periods.
User-friendly interface providing clear and detailed insights into the smart
home environment and agent behaviours
Specific required knowledge, skills, and/or technology: Proficiency in Python
and familiarity with the PyGame framework. Experience with deep learning,
particularly with large language models such as GPT-4. UI/UX design skills for
enhancing user interfaces. (Good to have but not essential).
Fields that this project may involve: Software Development; Artificial
Intelligence; NLP.
Resources provided by the client: Provide working project code as the starting
point. Provide detailed documentation as the current project description.
請(qǐng)加QQ：99515681 郵箱：99515681@qq.com WX：codinghelp
