In-game testing methods, devices, equipment, storage media, and program products

Through a collaborative architecture of intelligent agent model, large language model and in-game movement tool, automated testing of MOBA games is achieved, solving the problem that existing technologies cannot simulate the movement of virtual objects controlled by real players, and improving the effectiveness of performance testing and the authenticity of functional testing.

CN121560768BActive Publication Date: 2026-05-26TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
TENCENT TECHNOLOGY (SHENZHEN) CO LTD
Filing Date
2026-01-22
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing in-game testing methods for MOBA games cannot effectively simulate the movement of virtual objects by real players, resulting in performance tests lacking reference value and functional tests lacking realism.

Method used

Through a collaborative architecture of intelligent agent model, large language model and in-game movement tool, we achieve full-link automated testing of MOBA game in-game, simulating the movement process of real players controlling virtual objects, including the complete process of path planning, animation rendering and network synchronization.

Benefits of technology

It improves the effectiveness of performance testing and the authenticity of functional testing, provides a reliable basis for performance optimization, and reduces the complexity and maintenance cost of the testing system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121560768B_ABST
    Figure CN121560768B_ABST
Patent Text Reader

Abstract

This application discloses a method, apparatus, device, storage medium, and program product for in-game game testing. The method includes: inputting instruction text to an intelligent agent model to instruct in-game game testing; wherein different virtual objects participating in the game are controlled by different intelligent agent models; the intelligent agent model calls a large language model to perform task decomposition on the instruction text, obtaining a sequence of sub-tasks to be executed; when the intelligent agent model identifies a movement sub-task included in the sub-task sequence, it calls an in-game movement tool and inputs the position parameters of a target point to the in-game movement tool; the in-game movement tool obtains a path from the current position of the controlled object corresponding to the intelligent agent model to the target point based on the position parameters of the target point, and controls the controlled object to move along the path. This application significantly improves the effectiveness of in-game performance testing in MOBA games.
Need to check novelty before this filing date? Find Prior Art