Modeling An Agricultural Machine
A realistic dynamic simulation model of an agricultural machine can be developed using the multibody system dynamics approach. In this approach, several equations that can be solved in real time are employed to describe the dynamics of the machine. The dynamic equations may couple different engineering areas such as hydraulics and control systems. The real-time capability enables the solution of the complex equations in a time less than the simulation time-step of the model. This can provide users with a sense of continuously operating the machine without any computational glitches.
The literature offers several formulations to describe the multibody simulation of a working machine (Avello et a!., 1993; Bae, Han, &Yoo, 1999; Bayo, Jalon,& Serna, 1988; Bayo & Serna, 1989; Jalon, Alvarez, Ribera, Rodriguez, & Funes,2005). In this study, the simulation employs a semi-recursive formulation based on a velocity transformation (Avello et al., 1993), which results in a computationally efficient approach suitable for real-time applications. Figure 7.1 proposes a design process that can be used to facilitate a user-parameterized simulation model.
Design process for parameterization
When parameterizing a multibody model, a base model is made first, and the parameterized parts are provided as add-ons (Mohammadi, Kurvinen, & Mikkola, 2019). Model requirements such as geometries and the kinematic and dynamic properties of the machine are collected to build the base model. The
Figure 7Л A design algorithm to build a parameterized multibody model.
design specifications are collected for model parameterization. Depending on the number of parts to be parameterized, a separate file approach or a spreadsheet interface approach is taken. When the number of parameterized parts is less than or equal to five, for example, separate XML files are used along with the base model, which is also an XML file (Mohammadi, 2017). Otherwise, a spreadsheet interface is used, which is connected to the XML file of the base model using a script written in any programming language, such as Python (Jaiswal, 2017). Based on the parameterization requirements, the parts are assembled to carry out a feasibility analysis. If the feasibility analysis succeeds, the model is tested, and the design is iterated based on the model requirements.
The dynamic behavior of a customized real-time multibody model is dependent on the description of the virtual environment (Jaiswal et a!., 2019). To ensure a realistic dynamic simulation, it is essential to have an accurate description of the environment as well as the machine. Depending on simulation requirements, modeling certain environmental components makes it possible to represent the physical environment graphically. These components can contain complicated geometries and texture mapping. For this study of agricultural machinery, a deformable terrain environment can be modeled to facilitate the dynamic simulation in real time. This can be accomplished by combining mesh-based and particle-based methods, as already shown in Jaiswal et al. (2019).