Google filed 3 patents this quarter across 2 categories: AI & Machine Learning (2) and Networking (1).
The AI & Machine Learning patents cover systems for enhancing video game experiences, including automated discovery and creation of tutorials when players encounter difficulties, and an interactive coach that explains optimal gameplay decisions through conversational reasoning over recorded footage. The Networking patent addresses uplink scheduling optimization for 5G networks, specifically targeting reduced latency and jitter in extended reality and cloud gaming applications through dynamic resource allocation.
Google's 2 AI and machine learning patents both aim to help players improve at video games, but through different mechanisms. The first application describes a system that monitors where a player gets stuck during gameplay, then either searches for existing tutorials or automatically generates new ones from the player's own session recording if no suitable guide exists. The second patent takes a more conversational approach, allowing players to upload gameplay footage and ask an AI coach questions about why certain moves work better than others, with the system explaining its reasoning through back-and-forth dialogue rather than simply prescribing actions.
The Networking patent tackles a technical challenge in wireless extended reality and cloud gaming by giving devices more control over how they send data upstream to 5G networks. Instead of forcing devices to choose between pre-configured schedules (which are fast but inflexible) or requesting resources on demand (which is adaptable but slow), the application describes letting devices adjust their uplink transmission windows in real time within a pre-approved framework, reducing the lag and stuttering that plague wireless VR headsets and game Streaming.
All data sourced from USPTO patent filings. Google Patents may take several weeks to index recent publications. If a link is unavailable, search for the patent number at USPTO Patent Public Search.