Lecture 3, 3D scene modeling and simulation

During the development process of unmanned systems, a large number of semi-physical real-time simulation experiments of the motion control system are required. The motion trajectory display on the traditional simulation interface mostly adopts the form of two-dimensional curves, and the motion process and status are displayed through text data. As the amount of data increases, this method lacks intuition and is difficult to observe, understand and analyze, and is not conducive to mining the essential characteristics hidden in the data; while the information contained in a realistic image includes information displayed in curves and text. Therefore, applying three-dimensional visualization simulation technology to unmanned system simulation can display system data in a more natural and easier-to-understand image form, and realistically reproduce the entire operation process of the unmanned system.

The three-dimensional simulation of unmanned systems uses three-dimensional modeling technology to map the vehicle and its surrounding scenes into a virtual simulation environment (Visual Environment). Driven by the real-time data of the control computer, it truly reproduces the motion state of the vehicle and provides troubleshooting solutions. Provide strong analysis basis in detection, real-time monitoring, training simulation and other fields. In the three-dimensional simulation platform of unmanned systems, the key is to create a highly realistic simulation environment, and the authenticity and real-time nature of the display screen are the top priorities.

It is very challenging to ensure authenticity and real-time performance at the same time. For example, CG (Computer Graphics) technology in movies can render characters or scenes that are infinitely close to the real world. However, behind the scenes are a large number of high-performance computers that require long-term rendering and manual frame-by-frame rendering. The purpose of the current cutting-edge game development and virtual reality (Computer Graphics, VR) technology development is the same. The three-dimensional simulation of unmanned systems is expected to output an output that is as realistic as possible (authenticity) to the real world in the shortest possible time (real-time). At the same time, the three-dimensional simulation of unmanned systems also needs to be able to output simulation data of various visual sensors (camera, photoelectric pod, lidar, etc.) in real time, which requires higher computer performance and also relies on sensor modeling technology.

The RflySim platform mainly uses RflySim3D developed based on Unreal Engine for 3D simulation. In order to adapt to different simulation needs, the platform has built-in some scenes and 3D models, and provides corresponding external driver interfaces. If you need to use other scenes and 3D models, you should also Master the basic usage of some 3D processing software other than Unreal Engine. The interaction among RflySim3D and other software in the platform is realized through UDP, Python and Simulink control interfaces and script routines are provided.

• 3.1、How to use this chapter

• 3.2、CoreExps in this chapter

• 3.3、FAQs in this chapter

  Phase 2: Experimental Toolchain Configuration & 3D Scene Modeling and Simulation (Part 1)

  Phase II: Experimental Toolchain Configuration & 3D Scene Modeling and Simulation (Part 2)

  Phase 2: Experimental Toolchain Configuration & 3D Scene Modeling and Simulation (Part 3)