Utilization of simulations for the development of operator assistant systems in forest machines
Saaranen, Teemu (2023)
2023
Automaatiotekniikan DI-ohjelma - Master's Programme in Automation Engineering
Tekniikan ja luonnontieteiden tiedekunta - Faculty of Engineering and Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2023-10-24
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202310088683
https://urn.fi/URN:NBN:fi:tuni-202310088683
Tiivistelmä
Utilization of simulations for the development of operator assistant systems in forest machines are studied in this Master’s Thesis. The aim of this thesis is to create a basis for Ponsse’s operator assistant systems simulation by surveying and testing the suitability of different simulation software for Ponsse’s sensor simulation needs.
Several simulation tools and software options were surveyed for the simulations of operator assistant systems and the sensor simulation process provided by Mevea was chosen for a more detailed analysis. In this thesis, the creation process of the sensor simulation combination consisting of Mevea, Unity and ROS2 is clarified in detail and the suitability of this simulation process to Ponsse’s requirements is evaluated using relevant use cases.
In addition to Mevea’s sensor simulation process, the thesis also evaluates the suitability of alternative sensor simulation combinations for Ponsse’s needs. Alternative sensor simulation tools and software are Creanex’s sensor simulation combination in a training simulator and simulation software from Siemens, Cognata and CARLA for the development of advanced driver assistance systems and autonomous driving.
Based on the surveying of different software and simulation processes, the sensor simulation combination that meets Ponsse’s requirements should include a simulation process and a visual process. Game engines that provide immersive renderings require a suitable physics engine alongside them that the simulation combination can be utilized in practical R&D work and it brings enough benefits for Ponsse. The results of the thesis show that the game engine Unity offers a versatile and efficient tool for creating forest environments, developing autonomous systems and the possibility to integrate the physics of the forest machine into the simulation combination using built-in or external physics engine. Physics engine studies show that Mevea brings benefits for Ponsse with its advanced real-time physics engine which enables the development and testing of new features of forest machines. On the other hand, the physics engines and sensor simulation platforms focused on the automotive industry do not bring enough benefits for Ponsse’s forest industry needs.
Several simulation tools and software options were surveyed for the simulations of operator assistant systems and the sensor simulation process provided by Mevea was chosen for a more detailed analysis. In this thesis, the creation process of the sensor simulation combination consisting of Mevea, Unity and ROS2 is clarified in detail and the suitability of this simulation process to Ponsse’s requirements is evaluated using relevant use cases.
In addition to Mevea’s sensor simulation process, the thesis also evaluates the suitability of alternative sensor simulation combinations for Ponsse’s needs. Alternative sensor simulation tools and software are Creanex’s sensor simulation combination in a training simulator and simulation software from Siemens, Cognata and CARLA for the development of advanced driver assistance systems and autonomous driving.
Based on the surveying of different software and simulation processes, the sensor simulation combination that meets Ponsse’s requirements should include a simulation process and a visual process. Game engines that provide immersive renderings require a suitable physics engine alongside them that the simulation combination can be utilized in practical R&D work and it brings enough benefits for Ponsse. The results of the thesis show that the game engine Unity offers a versatile and efficient tool for creating forest environments, developing autonomous systems and the possibility to integrate the physics of the forest machine into the simulation combination using built-in or external physics engine. Physics engine studies show that Mevea brings benefits for Ponsse with its advanced real-time physics engine which enables the development and testing of new features of forest machines. On the other hand, the physics engines and sensor simulation platforms focused on the automotive industry do not bring enough benefits for Ponsse’s forest industry needs.