IP Library › Granted Patent US 12,528,016
Granted Patent B2
US 12,528,016 · App. 18/135,914 · Granted Jan 20, 2026

Artificial intelligence object control method and apparatus, device, and storage medium

Inventor: Shubao Zhang (Shenzhen, CN)
Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
A63F13/56G06F3/04815G06F3/04842G06F3/0486G06F3/0488G06T17/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,528,016
App. No.
18/135,914
Granted
Jan 20, 2026
Kind
B2
Abstract

Embodiments of this application disclose a method of controlling an artificial intelligence object performed by a terminal. The method includes: displaying, by the terminal, a virtual scene interface, the virtual scene interface including a virtual object controlled by the terminal; detecting, by the terminal, a control instruction via the virtual scene interface, the control instruction instructing an artificial intelligence object in an ally camp of the controlled virtual object to perform a target operation; and controlling, by the terminal based on the control instruction, the artificial intelligence object to perform the target operation.

Claims (84)

1 . A method of controlling an artificial intelligence object performed by a terminal and a sever that is communicatively connected to the terminal, the method comprising:

displaying, by the terminal, a virtual scene interface, the virtual scene interface including a virtual object controlled by the terminal and an artificial intelligence object in an ally camp of the controlled virtual object;

detecting, by the terminal, a control instruction from a user of the terminal via the virtual scene interface, the control instruction instructing the artificial intelligence object to perform a target operation;

transmitting, by the terminal, the control instruction to the server:

encoding, by the server, current status data of the artificial intelligence object through an encoding network to obtain encoded features of the artificial intelligence object;

performing, by the server, weighted fusion on the encoded features of the artificial intelligence object through a fusion network based on the control instruction to obtain a fused feature;

performing, by the server, operation prediction on the fused feature through an operation prediction network to predict operation information to be performed by the artificial intelligence object;

transmitting, by the server, the predicted operation information back to the terminal; and

controlling, by the terminal based on the control instruction, the artificial intelligence object to perform the target operation based on the predicted operation information in the virtual scene interface.

2 . The method according to claim 1 , wherein the detecting, by the terminal, the control instruction from the user of the terminal via the virtual scene interface comprises:

detecting, by the terminal, an attack instruction via the virtual scene interface, the attack instruction instructing the artificial intelligence object to attack a first target virtual object in a non-ally camp;

detecting, by the terminal, a retreat instruction via the virtual scene interface, the retreat instruction instructing the artificial intelligence object to retreat to a safe position; or

detecting, by the terminal, an assembling instruction via the virtual scene interface, the assembling instruction instructing the artificial intelligence object to move towards a position of the controlled virtual object or a second target virtual object of the ally camp.

3 . The method according to claim 2 , wherein the detecting, by the terminal, an attack instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on an attack control via the virtual scene interface, and generating the attack instruction, the first target virtual object being a virtual object closest to the controlled virtual object in the non-ally camp;

detecting, by the terminal, an operation of dragging the attack control to an object identifier of the first target virtual object via the virtual scene interface, and generating the attack instruction; or

detecting, by the terminal, a selection operation on the object identifier of the first target virtual object via the virtual scene interface, and generating the attack instruction.

4 . The method according to claim 2 , wherein the detecting, by the terminal, a retreat instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on a retreat control via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligence object to retreat to the safe position;

detecting, by the terminal, an operation of dragging the retreat control to an object identifier of a third target virtual object of the ally camp via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligent object to retreat to a position of the third target virtual object; or

detecting, by the terminal, a selection operation on the object identifier of the third target virtual object via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligence object to retreat to the position of the third target virtual object.

5 . The method according to claim 2 , wherein the detecting, by the terminal, an assembling instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on an assembling control via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the controlled virtual object;

detecting, by the terminal, an operation of dragging the assembling control to an object identifier of the second target virtual object of the ally camp via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the second target virtual object; or

detecting, by the terminal, a selection operation on the object identifier of the second target virtual object via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the second target virtual object.

6 . The method according to claim 1 , wherein the control instruction carries an object identifier of a target virtual object, and the target operation refers to an operation of moving towards the target virtual object; and

the controlling, by the terminal based on the control instruction, the artificial intelligence object to perform the target operation based on the predicted operation information in the virtual scene interface comprises:

controlling, by the terminal based on the control instruction, the artificial intelligence object to move towards the target virtual object.

7 . A computer game system including a terminal and a sever that is communicatively connected to the terminal, the terminal and the server each comprising a processor and a memory, the memory having at least one computer program stored therein, the at least one computer program being loaded and executed by the processor and causing the terminal and the server to implement a method of controlling an artificial intelligence object including:

displaying, by the terminal, a virtual scene interface, the virtual scene interface including a virtual object controlled by the terminal and an artificial intelligence object in an ally camp of the controlled virtual object;

detecting, by the terminal, a control instruction from a user of the terminal via the virtual scene interface, the control instruction instructing the artificial intelligence object to perform a target operation;

transmitting, by the terminal, the control instruction to the server;

encoding, by the server, current status data of the artificial intelligence object through an encoding network to obtain encoded features of the artificial intelligence object;

performing, by the server, weighted fusion on the encoded features of the artificial intelligence object through a fusion network based on the control instruction to obtain a fused feature;

performing, by the server, operation prediction on the fused feature through an operation prediction network to predict operation information to be performed by the artificial intelligence object;

transmitting, by the server, the predicted operation information back to the terminal; and

controlling, by the terminal based on the control instruction, the artificial intelligence object to perform the target operation based on the predicted operation information in the virtual scene interface.

8 . The computer game system according to claim 7 , wherein the detecting, by the terminal, the control instruction from the user of the terminal via the virtual scene interface comprises:

detecting, by the terminal, an attack instruction via the virtual scene interface, the attack instruction instructing the artificial intelligence object to attack a first target virtual object in a non-ally camp;

detecting, by the terminal, a retreat instruction via the virtual scene interface, the retreat instruction instructing the artificial intelligence object to retreat to a safe position; or

detecting, by the terminal, an assembling instruction via the virtual scene interface, the assembling instruction instructing the artificial intelligence object to move towards a position of the controlled virtual object or a second target virtual object of the ally camp.

9 . The computer game system according to claim 8 , wherein the detecting, by the terminal, an attack instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on an attack control via the virtual scene interface, and generating the attack instruction, the first target virtual object being a virtual object closest to the controlled virtual object in the non-ally camp;

detecting, by the terminal, an operation of dragging the attack control to an object identifier of the first target virtual object via the virtual scene interface, and generating the attack instruction; or

detecting, by the terminal, a selection operation on the object identifier of the first target virtual object via the virtual scene interface, and generating the attack instruction.

10 . The computer game system according to claim 9 , wherein the detecting, by the terminal, a retreat instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on a retreat control via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligence object to retreat to the safe position;

detecting, by the terminal, an operation of dragging the retreat control to an object identifier of a third target virtual object of the ally camp via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligent object to retreat to a position of the third target virtual object; or

detecting, by the terminal, a selection operation on the object identifier of the third target virtual object via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligence object to retreat to the position of the third target virtual object.

11 . The computer game system according to claim 8 , wherein the detecting, by the terminal, an assembling instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on an assembling control via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the controlled virtual object;

detecting, by the terminal, an operation of dragging the assembling control to an object identifier of the second target virtual object of the ally camp via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the second target virtual object; or

detecting, by the terminal, a selection operation on the object identifier of the second target virtual object via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the second target virtual object.

12 . The computer game system according to claim 7 , wherein the control instruction carries an object identifier of a target virtual object, and the target operation refers to an operation of moving towards the target virtual object; and

the controlling, by the terminal based on the control instruction, the artificial intelligence object to perform the target operation based on the predicted operation information in the virtual scene interface comprises:

controlling, by the terminal based on the control instruction, the artificial intelligence object to move towards the target virtual object.

13 . A non-transitory computer-readable storage medium, having at least one computer program stored therein, the at least one computer program being loaded and executed by a computer game system including a terminal and a sever that is communicatively connected to the terminal and causing the terminal and the server to implement a method of controlling an artificial intelligence object including:

displaying, by the terminal, a virtual scene interface, the virtual scene interface including a virtual object controlled by the terminal and an artificial intelligence object in an ally camp of the controlled virtual object;

detecting, by the terminal, a control instruction from a user of the terminal via the virtual scene interface, the control instruction instructing the artificial intelligence object to perform a target operation;

transmitting, by the terminal, the control instruction to the server;

encoding, by the server, current status data of the artificial intelligence object through an encoding network to obtain encoded features of the artificial intelligence object;

performing, by the server, weighted fusion on the encoded features of the artificial intelligence object through a fusion network based on the control instruction to obtain a fused feature;

performing, by the server, operation prediction on the fused feature through an operation prediction network to predict operation information to be performed by the artificial intelligence object;

transmitting, by the server, the predicted operation information back to the terminal; and

controlling, by the terminal based on the control instruction, the artificial intelligence object to perform the target operation based on the predicted operation information in the virtual scene interface.

14 . The non-transitory computer-readable storage medium according to claim 13 , wherein the detecting, by the terminal, the control instruction from the user of the terminal via the virtual scene interface comprises:

detecting, by the terminal, an attack instruction via the virtual scene interface, the attack instruction instructing the artificial intelligence object to attack a first target virtual object in a non-ally camp;

detecting, by the terminal, a retreat instruction via the virtual scene interface, the retreat instruction instructing the artificial intelligence object to retreat to a safe position; or

detecting, by the terminal, an assembling instruction via the virtual scene interface, the assembling instruction instructing the artificial intelligence object to move towards a position of the controlled virtual object or a second target virtual object of the ally camp.

15 . The non-transitory computer-readable storage medium according to claim 14 , wherein the detecting, by the terminal, an attack instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on an attack control via the virtual scene interface, and generating the attack instruction, the first target virtual object being a virtual object closest to the controlled virtual object in the non-ally camp;

detecting, by the terminal, an operation of dragging the attack control to an object identifier of the first target virtual object via the virtual scene interface, and generating the attack instruction; or

detecting, by the terminal, a selection operation on the object identifier of the first target virtual object via the virtual scene interface, and generating the attack instruction.

16 . The non-transitory computer-readable storage medium according to claim 14 , wherein the detecting, by the terminal, a retreat instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on a retreat control via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligence object to retreat to the safe position;

detecting, by the terminal, an operation of dragging the retreat control to an object identifier of a third target virtual object of the ally camp via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligent object to retreat to a position of the third target virtual object; or

detecting, by the terminal, a selection operation on the object identifier of the third target virtual object via the virtual scene interface, and generating the retreat instruction, the retreat instruction being used for instructing the artificial intelligence object to retreat to the position of the third target virtual object.

17 . The non-transitory computer-readable storage medium according to claim 14 , wherein the detecting, by the terminal, an assembling instruction via the virtual scene interface comprises:

detecting, by the terminal, a trigger operation on an assembling control via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the controlled virtual object;

detecting, by the terminal, an operation of dragging the assembling control to an object identifier of the second target virtual object of the ally camp via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the second target virtual object; or

detecting, by the terminal, a selection operation on the object identifier of the second target virtual object via the virtual scene interface, and generating the assembling instruction, the assembling instruction being used for instructing the artificial intelligence object to move towards the position of the second target virtual object.

18 . The non-transitory computer-readable storage medium according to claim 13 , wherein the control instruction carries an object identifier of a target virtual object, and the target operation refers to an operation of moving towards the target virtual object; and

the controlling, by the terminal based on the control instruction, the artificial intelligence object to perform the target operation based on the predicted operation information in the virtual scene interface comprises:

controlling, by the terminal based on the control instruction, the artificial intelligence object to move towards the target virtual object.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 18, 2023
From: ZHANG, SHUBAO
To: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
Reel/Frame 063358/0324 →
Priority Claims (1)
CN 202110969023.3 · Aug 23, 2021 · national
Continuity (2)
Continuation PCTCN2022106330 · Jul 18, 2022
Related Publication 20230293995A1 · Sep 21, 2023
References Cited (23)
US 9827492B2 · Yamada · 2017 [cited by examiner]
US 10864446B2 · Borovikov · 2020 [cited by examiner]
US 11471764B2 · Kolen · 2022 [cited by examiner]
US 20210158141A1 · Taylor et al. · 2021 [cited by applicant]
US 20220274023A1 · Vange · 2022 [cited by examiner]
US 20230173396A1 · Asai · 2023 [cited by examiner]
US 20230219000A1 · Dang · 2023 [cited by examiner]
US 20230256338A1 · Cai · 2023 [cited by examiner]
US 20230256341A1 · Cai · 2023 [cited by examiner]
CN 110310350A · 2019 [cited by applicant]
CN 111282279A · 2020 [cited by applicant]
CN 111450533A · 2020 [cited by applicant]
CN 111514585A · 2020 [cited by applicant]
CN 112973117A · 2021 [cited by applicant]
CN 113254872A · 2021 [cited by applicant]
CN 114272599A · 2022 [cited by applicant]
WO WO2020252400A1 · 2020 [cited by applicant]
Tencent Technology, WO, PCT/CN2022/106330, Oct. 19, 2022, 5 pgs. [cited by applicant]
Tencent Technology, IPRP, PCT/CN2022/106330, Feb. 27, 2024, 6 pgs. [cited by applicant]
Deep Learning and Advanced Intelligent Decision Making, “Tencent's “Juewu” Paper Discloses Technical Details”, Mar. 25, 2020, 5 pgs., Retrieved from the Internet: https://aljishu.com/a/1060000000100723. [cited by applicant]
Tencent Technology, ISR, PCT/CN2022/106330, Oct. 19, 2022, 3 pgs. [cited by applicant]
Tieba. Baidu.com, “How to Control the Actions of AI Teammates?”, Apr. 2018, Retrieved from the Internet: https://tieba.baidu.com/p/5634743473. [cited by applicant]
Zhuanlan.Zhihu.com, “Playing MOBA Games Using Deep Reinforcement Learning”, Dec. 2020, Retrieved from the Internet: https://zhuanlan.zhihu.com/p/378789632. [cited by applicant]