A661Widget drawing method based on adaptive depth caching

The drawing process of A661Widget is optimized by using an adaptive depth caching method, which solves the problem of low primitive drawing efficiency in complex UA scenarios and achieves high efficiency in primitive drawing and rendering.

CN116612208BActive Publication Date: 2026-11-10CHINESE AERONAUTICAL RADIO ELECTRONICS RES INST
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202310622592.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-05-30
Publication Date
2026-11-10
Estimated Expiration
2043-05-30

AI Technical Summary

Technical Problem

In complex user agent (UA) scenarios, existing technologies cannot efficiently render a large number of A661 primitives, resulting in excessively long rendering times and failing to meet the requirements for high-performance rendering.

Method used

An adaptive depth caching method is adopted, which calculates depth parameters through an adaptive UA algorithm, rationally allocates depth buffer pool resources, optimizes the drawing order and information transmission of primitives, and reduces the number of primitive vertex transmissions.

Benefits of technology

Despite the limited hardware performance of the IDU, the efficiency of primitive drawing was improved, the drawing time was reduced, and the rendering efficiency of the cockpit display system was enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116612208B_ABST
    Figure CN116612208B_ABST
Patent Text Reader

Abstract

This invention discloses an A661Widget rendering method based on adaptive depth caching. It estimates the maximum theoretical limit of primitives based on the definition files of each user application and then sets an appropriate depth step value. According to the Widget rendering order in the DF file, depth values ​​are assigned to the basic primitives of all Widget components in each cycle according to the depth step value. Basic primitives not yet rendered in the current cycle are placed in a primitive depth buffer pool to ensure that batch-rendered primitives meet the rendering order requirements of the User Agent (UA). During Widget rendering, the vertex coordinate information, color information, and texture coordinates of the primitives are organized. When the number of primitives in the primitive depth buffer pool reaches its limit or the current rendering cycle ends, the vertex rendering information of the primitives is uniformly transmitted to the rendering pipeline. This invention significantly increases the number of primitives rendered per unit time in the display system, resulting in a marked improvement in the rendering efficiency of the display system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of integrated avionics system design, and specifically relates to an A661Widget drawing method based on adaptive depth caching. Background Technology

[0002] As avionics systems become increasingly integrated, the content displayed in cockpit displays is also expanding. The ARINC 661 specification divides the cockpit display system into two functional modules: User Application (UA) and Display System (CDS), aiming to completely isolate the interface's display functions from its logical processing functions. The UA module handles the logical processing and sends the drawing requirements to the CDS module. The CDS module then draws the A661 Widget components for each UA layer according to the A661 specification. These A661 Widget components must draw various primitives as needed, including basic primitives such as lines, text, and polygons.

[0003] Basic primitives are used in the display requirements of various User Agent (UA) applications, such as flight management and main unit display applications. The main characteristics of basic primitives are: 1) Each type of basic primitive has different shape, size, color, and other primitive information, meaning a large number of vertices need to be transmitted to the drawing pipeline. 2) The number of primitives in each type is large, and each UA application needs to draw a large number of basic primitives. 3) The drawing order of primitives is affected by the real-time UA, and the drawing order of each primitive cannot be obtained in advance. 4) If drawing is done by transmitting the vertex information of each primitive to the drawing pipeline after it is drawn, it will result in a large amount of drawing time consumption, making it impossible to meet the high-performance drawing requirements of complex UA scenarios. Summary of the Invention

[0004] To address the high-performance requirements of rendering large batches of A661 primitives in complex User Agent (UA) scenarios, this invention aims to provide an A661Widget rendering method based on adaptive depth caching. By assigning appropriate depth values ​​to the Widget components of each A661 layer, it avoids batch rendering of Widget component primitives, enabling large-scale primitive rendering. Simultaneously, an adaptive UA algorithm automatically calculates the depth parameter value to meet the primitive rendering quantity requirements caused by different UAs. This invention supports both meeting primitive rendering order requirements and ensuring that UA layers with complex content but high real-time requirements can be rendered quickly, improving the high-performance performance of A661 rendering functionality on various complex UAs.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A drawing method for A661Widget based on adaptive depth caching includes the following steps:

[0007] Adaptive depth parameters: Estimate the maximum theoretical limit number of primitives based on the definition files of each user application, and then set an appropriate depth step value;

[0008] Depth Allocation and Caching: Based on the Widget drawing order in the DF file, the depth value is assigned to the basic primitives of all Widget components according to the depth step value in each cycle. The basic primitives that are not drawn in the current cycle are put into the primitive depth buffer pool to ensure that the primitives drawn in batches meet the drawing order requirements of the UA.

[0009] Display vertex organization and transmission: When a widget is drawn, the vertex coordinate information, color information, and texture coordinates of primitives are organized. When the number of primitives in the primitive depth buffer pool reaches the upper limit or the current drawing cycle ends, the vertex drawing information of the primitives is uniformly transmitted to the rendering pipeline.

[0010] Preferably, the process of calculating the depth stride value includes the following steps:

[0011] 1) Read the DF file of the UA and read the widget components contained in each layer;

[0012] 2) Analyze the maximum number of primitives that need to be drawn during the drawing process of each widget;

[0013] 3) Sum the maximum number of primitives for each widget in the layer to get the maximum possible number of primitives for the layer, which is the theoretical limit number of primitives for that layer;

[0014] 4) Sum the theoretical limit number of primitives for each layer to obtain the maximum theoretical limit number of primitives, MaxGpNum;

[0015] 5) Set the initial value of the UA depth value, and set the depth step size to the initial value / MaxGpNum.

[0016] Preferably, the detailed steps for depth allocation and caching, display vertex organization and transmission are as follows:

[0017] 1) Clear the depth buffer pool of all types of primitives;

[0018] 2) Execute the drawing functions for each widget within the layer;

[0019] 3) The drawing functions of each widget execute the drawing functions of various graphic elements;

[0020] 4) Global depth value minus depth step size;

[0021] 5) Set the current primitive's depth value to the global depth value, and record the color and coordinate information into the depth buffer pool;

[0022] 6) Repeat step 3 until all widget drawing functions in the current layer have finished executing;

[0023] 7) Transfer the vertex array of this layer to the OpenGL rendering pipeline;

[0024] 8) The global depth value is restored to its initial value.

[0025] The beneficial effects of this invention are as follows:

[0026] This invention, under the condition of limited IDU hardware performance, reduces the number of transmissions of primitive vertices by introducing adaptive UA depth into primitives, thereby reducing the overall drawing time of Widget components and significantly increasing the number of primitives drawn per unit time in the cockpit display system. Attached Figure Description

[0027] Figure 1 A schematic diagram of the drawing process of the primitive interface layer.

[0028] Figure 2 Adaptive depth parameter module diagram.

[0029] Figure 3 Schematic diagram of depth allocation and caching module.

[0030] Figure 4 Displaying the vertex organization and transmission module diagram. Detailed Implementation

[0031] All English terms involved in this invention are defined in the ARINC 661 standard.

[0032] The commonly used technical terms in this field involved in this invention are shown in the following table:

[0033] Technical terms English Chinese CDS Cockpit Display System Cockpit Display System UA User Application User Application DF Define File Definition file OpenGL Open Graphics Library Open Graphics Library

[0034] To better understand the present invention, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0035] like Figure 1 As shown, during the process of merging and drawing multiple primitives, since the primitive interface layer cannot know the drawing order of each primitive, when each type of primitive interface is called, regardless of the number of primitives received, in order to ensure the correctness of the final image drawing order, it is necessary to directly pass the primitive information of the same type to the basic graphics library, and then the basic graphics library calls OpenGL to draw. This makes the total drawing time proportional to the number of times the primitive interface layer is called, thus affecting the drawing efficiency.

[0036] The A661Widget drawing method based on adaptive depth caching shown in this embodiment runs on CDS and includes the following functional modules: adaptive depth parameter module, depth allocation and caching module, and display vertex organization and transmission module.

[0037] Adaptive Depth Parameter Module: When multiple User Agent (UA) applications run simultaneously within an Integrated Display Unit (IDU), the UA's Definition File (DF) defines the layer order and the parameters for defining widget components within each layer. Because the number of primitives that different UA applications need to draw varies significantly, a single set of depth value parameters cannot be used. The adaptive depth parameter module estimates the maximum theoretical limit of primitives based on the DF files of each UA application and then sets an appropriate depth step value. Through the depth step value, the depth value parameters can be adaptively adjusted for the actual situation of different UA applications within the same window, rationally allocating depth buffer pool resources to meet the different primitive drawing requirements of each UA, such as... Figure 2 As shown, the detailed steps of the adaptive depth parameter module are as follows:

[0038] 1) Read the DF file of the UA and read the widget components contained in each layer;

[0039] 2) Analyze the maximum number of primitives that may need to be drawn during the drawing process of each widget;

[0040] 3) Sum the maximum number of primitives for each widget in the layer to get the maximum possible number of primitives for the layer, which is the theoretical limit number of primitives for that layer;

[0041] 4) Sum the theoretical limit number of primitives for each layer to obtain the maximum theoretical limit number of primitives, MaxGpNum;

[0042] 5) Set the initial depth value of the UA to 0.5 and the depth step value to 0.5 / MaxGpNum.

[0043] Depth Allocation and Caching Module: Based on the Widget drawing order in the DF file, the module assigns depth values ​​to the basic primitives of all Widget components according to the depth step value within each cycle. Undrawn basic primitives in the current cycle are placed in the primitive depth buffer pool to ensure that batch-drawn primitives meet the UA's drawing order requirements. The A661 software of this invention introduces primitive depth information into the traditional two-dimensional coordinate system of primitives. In the x,y coordinate system, each primitive is marked with its own drawing depth value according to the Widget drawing order in the DF file. Since OpenGL's coordinate system is a right-handed coordinate system, it is only necessary to set the depth value of the primitive drawn first to be greater than the depth value of the primitive drawn later. Then, at the end of each A661 layer, all vertex array data information of this layer is rendered, achieving primitive drawing while satisfying the drawing order.

[0044] The display vertex organization and transmission module organizes vertex coordinate information, color information, texture coordinates and other information of primitives when the A661 Widget is drawn. When the number of primitives in the primitive depth buffer pool reaches the upper limit or the current drawing cycle ends, the vertex drawing information of the primitives is uniformly transmitted to the rendering pipeline.

[0045] like Figure 3 As shown, the detailed steps of the depth allocation and caching module and the display vertex organization and transmission module are as follows:

[0046] 1) Clear the depth buffer pool of all types of primitives;

[0047] 2) Execute the drawing functions for each widget within the layer;

[0048] 3) The drawing functions of each widget execute the drawing functions of various graphic elements;

[0049] 4) Subtract the depth step size value of 0.5 / MaxGpNum from the global depth value;

[0050] 5) Set the depth value of the current primitive to the global depth value, such as... Figure 4 As shown, the vertex organization and transmission module records color and coordinate information into the depth buffer pool;

[0051] 6) Repeat step 3 until all widget drawing functions in the current layer have finished executing;

[0052] 7) The vertex organization and transmission module is responsible for transmitting the vertex array of this layer to the OpenGL rendering pipeline;

[0053] 8) The global depth value is restored to 0.5.

[0054] To address the issue of low rendering efficiency of primitive layers while adhering to rendering order and adapting to various rendering user agent (UA) application scenarios, this invention proposes an A661 depth management rendering optimization scheme using OpenGL 3D primitives. This significantly increases the number of primitives rendered per unit time in the display system, resulting in a marked improvement in the rendering efficiency of the display system.

[0055] It is understood that those skilled in the art can make equivalent substitutions or modifications to the technical solution and inventive concept of the present invention, and all such substitutions or modifications should fall within the protection scope of the appended claims.

Claims

1. A method for drawing A661Widgets based on adaptive depth caching, characterized in that... Includes the following steps: Adaptive Depth Parameters: Based on the definition file (DF) of each user application (UA), the maximum theoretical limit number of primitives is estimated, and then an appropriate depth step value is set; the calculation process of the depth step value includes the following steps: 1) Read the DF of the UA and read the widget components contained in each layer; 2) Analyze the maximum number of primitives that need to be drawn during the drawing process of each widget; 3) Sum the maximum number of primitives for each widget in the layer to get the maximum possible number of primitives for the layer, which is the theoretical limit number of primitives for that layer; 4) Sum the theoretical limit number of primitives for each layer to obtain the maximum theoretical limit number of primitives, MaxGpNum; 5) Set the initial value of the UA depth value, and set the depth step size to the initial value / MaxGpNum; Depth Allocation and Caching: Based on the drawing order of Widgets in DF, the depth value is assigned to the basic primitives of all Widget components according to the depth step value in each cycle. The basic primitives that are not drawn in the current cycle are put into the primitive depth buffer pool to ensure that the primitives drawn in batches meet the drawing order requirements of UA. Display vertex organization and transmission: When a widget is drawn, the vertex coordinate information, color information, and texture coordinates of primitives are organized. When the number of primitives in the primitive depth buffer pool reaches the upper limit or the current drawing cycle ends, the vertex drawing information of the primitives is uniformly transmitted to the rendering pipeline.

2. The A661Widget drawing method based on adaptive depth caching according to claim 1, characterized in that... The detailed steps for depth allocation and caching, and display vertex organization and transmission are as follows: 1) Clear the depth buffer pool of all types of primitives; 2) Execute the drawing functions for each widget within the layer; 3) The drawing functions of each widget execute the drawing functions of various graphic elements; 4) Global depth value minus depth step size; 5) Set the current primitive's depth value to the global depth value, and record the color and coordinate information into the depth buffer pool; 6) Repeat step 3 until all widget drawing functions in the current layer have finished executing; 7) Transfer the vertex array of the current layer to the OpenGL rendering pipeline; 8) The global depth value is restored to its initial value.

Citation Information

Patent Citations

  • Instruction filtering method and device for specifications of ARINC661

    CN104331278A

  • Fuzzy control creation method and device and computing equipment

    CN114756323A