Partitioned AI for Networked Games

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current gaming servers lack sufficient computing power to support sophisticated artificial intelligence for networked games, leading to unchallenging AI behavior in bots, due to limitations in computational resources and latency issues when offloading computations to clients.

Innovation Solution

Implementing a partitioned artificial intelligence system that splits AI into a lightweight server-side component and a computationally intensive client-side component, using a Taylor-series approximation to calculate influence fields, allowing for enhanced game realism without radical architectural changes and addressing latency and client failure concerns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If AI computations are performed centrally on the gaming server, then game state management is centralized and consistent, but the server lacks sufficient computing power to support sophisticated AI for multiple bots

Engineering Contradiction:

Improvegame state consistencyVSAvoidAI computational capacity

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent divides AI computation into two segments: a lightweight server-side component that maintains game state consistency and a client-side component that performs computationally intensive AI calculations. This segmentation allows sophisticated AI to run on clients while the server maintains authoritative game state through periodic synchronization.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a new architectural dimension by moving AI computation from the traditional single-server model to a distributed client-server model. Clients become active participants in AI computation, transforming the centralized computation model into a distributed one that leverages aggregate client computing power.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Productivity

If AI computations are offloaded to clients, then computational power is increased, but communication latency between clients and server increases

Engineering Contradiction:

ImproveAI computational capacityVSAvoidround-trip latency

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent implements preliminary action by having clients perform AI computations in advance and send results to the server before they are needed for game state updates. This allows the server to receive pre-computed AI results, reducing the effective latency impact on game proceedings.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the balance between client-side computation and server-side validation based on network conditions and game state requirements. During periods of high latency, the system relies more on client-side autonomy, while during low-latency periods, synchronization occurs more frequently.

Inventive Principle:

Principle #15Dynamics

3Productivity

If AI computations are offloaded to clients, then computational power is increased, but client machines may fail or be compromised

Engineering Contradiction:

ImproveAI computational capacityVSAvoidcomputation availability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent introduces a lightweight server-side AI component as an intermediary that validates and supervises client-side AI computations. This intermediary ensures that even if clients fail or are compromised, the server can detect inconsistencies and maintain game integrity through periodic verification and authoritative game state management.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Productivity

If more server resources are added to support sophisticated AI, then AI intelligence is improved, but operational costs become prohibitive

Engineering Contradiction:

ImproveAI computational capacityVSAvoidserver operational cost

Core Design Contradiction:

ProductivityVSUse of energy by stationary object

Solution Approach 1:

The patent implements self-service by enabling client machines to perform AI computations for themselves and other players in the game. Instead of requiring the server to provide all computational resources, clients serve their own AI computation needs using their local hardware, dramatically reducing server resource requirements and operational costs.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS9327194B2Partitioned artificial intelligence for networked games

Publication Date: 2016.05.03 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Partitioned artificial intelligence (AI) for networked gaming. An exemplary system splits the AI into a computationally lightweight server-side component and a computationally intensive client-side component to harness the aggregate computational power of numerous gaming clients. Aggregating resources of many, even thousands of client machines enhances game realism in a manner that would be prohibitively expensive on the central server. The system is tolerant of latency between server and clients. Deterministic and stateless client-side components enable rapid handoff, preemptive migration, and replication of the client-side AI to address problems of client failure and game exploitation. The partitioned AI can support tactical gaming navigation, a challenging task to offload because of sensitivity to latency. The tactical navigation AI calculates influence fields partitioned into server-side and client-side components by means of a Taylor-series approximation.