A 3D model can bring realism to a simulation of a scene, and it also helps to identify everything in the scene more easily. For example, if all objects in a scene are simple shapes, such as cubes and circles, it can be difficult to distinguish objects in simulation.
1. Collision shape and visual shape
Simple shapes like cubes and spheres, often called "primitives", are often used as collision blocks. Collision shapes use simplified geometry compared to visual meshes because they serve different purposes in simulations.
Collision models are important because they interact with the physical environment and are less computationally expensive to use basic shapes.
Visual shapes are detailed meshes. The purpose of making these shapes is to bring realism to the simulated scene. These shapes are denser in polygons and often have textures attached to them. These shapes are picked up by the camera sensor and are what the viewer sees when viewing the simulation.
2. Create a 3D model
Blender has many tools to create simple/complex models. A good way to create complex objects is to start with simple things. In the example below we see a complex model created from simple primitives. While this is a good modeling strategy, going from a simple mesh to a complex mesh often requires the use of many tools.
Since we used the wheel example above, I will use that shape for my modeling workflow. I'll also explain some of the tools and modifiers I use. Before starting this tutorial, some basic understanding of Blender hotkeys is recommended.