# How AI Learns to Control Game Characters Based on Their Surroundings > A system that allows digital characters to automatically perform actions by matching their current environment to previously learned experiences stored in a database. - **Patent:** US 10607134 - **Original title:** Artificially intelligent systems, devices, and methods for learning and/or using an avatar's circumstances for autonomous avatar operation - **Owner:** Individual - **Granted:** 2020 - **Status:** Active - **Times cited:** 26 - **Field:** gaming, ai_ml, software ## What it does This patent describes a method for teaching a digital character, or avatar, to act on its own by recognizing patterns in its environment. The system maintains a knowledgebase that links specific environmental objects to sets of instructions or actions. When the avatar encounters a new situation, the system compares the current objects in the scene to the stored patterns. If a match is found, the system triggers the corresponding action, allowing the avatar to navigate or interact with the game world without manual player input. ## What it does NOT cover - Does not cover manual control of avatars by human players. - Does not cover basic scripted AI behaviors that are hard-coded rather than learned via pattern matching. - Does not cover the underlying physics engines used to render the game objects themselves. ## The clever bit The system uses a correlation-based knowledgebase that treats environmental objects as data points, allowing the AI to transfer learned behaviors from one avatar or application to another if the environmental context matches. ## Real-world examples 1. Autonomous NPCs in open-world role-playing games 2. Automated testing bots for game quality assurance 3. AI-driven character training in game development environments ## Why it matters As games become more complex, manual programming for every possible NPC (non-player character) behavior is inefficient. This approach moves toward autonomous agents that can adapt to different game states, which is a core challenge in modern game development and simulation software. ## Frequently asked questions ### What does How AI Learns to Control Game Characters Based on Their Surroundings cover? A system that allows digital characters to automatically perform actions by matching their current environment to previously learned experiences stored in a database. ### Who owns patent US 10607134? Individual owns this patent, granted in 2020. ### When does this patent expire? This patent is expected to expire on December 19, 2036, when the invention enters the public domain. ### What is patent US 10607134 cited by? This patent has been cited by 26 later patents that build on its ideas. ### What problem does this patent solve? As games become more complex, manual programming for every possible NPC (non-player character) behavior is inefficient. This approach moves toward autonomous agents that can adapt to different game states, which is a core challenge in modern game development and simulation software. ### What does this patent NOT cover? Does not cover manual control of avatars by human players. **Full plain-English explainer:** https://patentbrief.org/patent/us/10607134/artificially-intelligent-systems-devices-and-methods-for-learning-andor-using-an-avatars-circumstances-for-autonomous-avatar-operation **Original patent:** https://patents.google.com/patent/US10607134 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How AI Agents Learn to Pick the Best Future Actions](https://patentbrief.org/patent/us/10282665/action-selection-with-a-reward-estimator-applied-to-machine-learning) — A method for an AI agent to predict which actions will yield the highest rewards by analyzing past experiences and refining its decision-making model. - [Nvidia's Method for Training Self-Driving Car AI in Simulations](https://patentbrief.org/patent/us/11436484/alphazero) — Nvidia's 2022 patent describes how to train AI for self-driving cars by using simulated environments and virtual sensors, then matching the simulated data format to real-world sensor data for AI processing. - [How AI Systems Learn to Predict and Act Simultaneously](https://patentbrief.org/patent/us/11170293/alphago-policy-and-value-networks) — A method for training AI models that combines supervised learning for prediction with reinforcement learning for decision-making in a single, coordinated system. - [How AI Agents Learn Faster by Predicting Opponents in Parallel](https://patentbrief.org/patent/us/11295174/opponent-modeling-with-asynchronous-methods-in-deep-rl) — This patent describes a computer system that trains artificial intelligence agents more efficiently by using many processors to simultaneously learn from a simulated environment, while also explicitly predicting the actions of other agents. - [How AI Learns to Manage Computer Applications Using Digital Simulators](https://patentbrief.org/patent/us/11238372/simulator-training-for-automated-reinforcement-learning-based-application-manage) — This patent describes a system that creates and trains digital simulators, which then teach other AI programs how to automatically manage and optimize complex computer applications in a controlled virtual environment.