NPC Behavior Tree Selection Based on Player Distance

Solution Overview

Problem

In open world games, a large number of AI objects using the same complex behavior tree lead to excessive server resource consumption, compromising performance and player experience.

Innovation Solution

A method where the distance between NPCs and player-controlled objects determines the complexity of the behavior tree used, with closer NPCs employing more resource-intensive trees for rich representations and farther NPCs using less resource-intensive trees, maintaining a balance in server resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If a complex behavior tree is used to achieve rich NPC representations, then the NPC representation quality is improved, but the server resource consumption increases

Engineering Contradiction:

ImproveNPC representation qualityVSAvoidserver resource consumption

Core Design Contradiction:

Manufacturing precisionVSUse of energy by moving object

Solution Approach 1:

The patent applies local quality by differentiating behavior tree complexity based on NPC distance to player. High-quality complex behavior trees are used only for NPCs close to players (within threshold distance), while simpler behavior trees are used for distant NPCs. This ensures rich representations where needed without uniformly high resource consumption across all NPCs.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent implements dynamics by making the behavior tree selection adaptive and dynamic. The system continuously monitors NPC-player distances and dynamically switches between different behavior tree complexities based on real-time spatial relationships, allowing the system to optimize resource allocation as NPCs move relative to players.

Inventive Principle:

Principle #15Dynamics

2Stability of the object's composition

If the same complex behavior tree is used for all AI objects, then the NPC representation consistency is improved, but the server performance deteriorates

Engineering Contradiction:

ImproveNPC representation consistencyVSAvoidserver performance

Core Design Contradiction:

Stability of the object's compositionVSPower

Solution Approach 1:

The patent applies local quality by differentiating behavior tree complexity based on NPC distance to player. High-quality complex behavior trees are used only for NPCs close to players (within threshold distance), while simpler behavior trees are used for distant NPCs. This ensures rich representations where needed without uniformly high resource consumption across all NPCs.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent applies partial action by using complex behavior trees only partially - specifically for NPCs within the threshold distance of players. For NPCs beyond this distance, the system uses simplified behavior trees. This partial application of complex behavior maintains representation quality where most impactful (near player view) while reducing overall server load.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If more AI objects are added to the virtual environment, then the game world richness is improved, but the server resource consumption increases

Engineering Contradiction:

Improvegame world richnessVSAvoidserver resource consumption

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent applies local quality by differentiating behavior tree complexity based on NPC distance to player. High-quality complex behavior trees are used only for NPCs close to players (within threshold distance), while simpler behavior trees are used for distant NPCs. This ensures rich representations where needed without uniformly high resource consumption across all NPCs.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent uses copying by creating multiple behavior tree variants (first behavior tree with high complexity, second behavior tree with low complexity) that can be selectively copied and applied to different NPCs based on their spatial relationship with players, rather than using a single universal behavior tree for all NPCs.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS20230302362A1Virtual object control method based on distance from player-controlled virtual object

Publication Date: 2023.09.28 TENCENT TECHNOLOGY (SHENZHEN) CO LTD

AI summary

A virtual object control method includes controlling a first virtual object in a virtual environment through a first behavior tree, the first virtual object being a Non-Player character (NPC) in the virtual environment, and obtaining a first distance between the first virtual object and a second virtual object controlled by a player in the virtual environment, the second virtual object being a closest virtual object controlled by a player to the first virtual object. The method further includes selecting, based on the first distance, a behavior tree among plural behavior trees for controlling the first virtual object. The first distance inversely relates to an amount of computing resources consumed during use of the selected behavior tree.