Project Details
Projekt Print View

Combined Structural and Dimensional Synthesis of Task Redundant Robot Manipulators

Applicant Dr.-Ing. Svenja Spindeldreier, since 5/2021
Subject Area Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics
Term from 2017 to 2022
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 341489206
 
Final Report Year 2023

Final Report Abstract

In the completed project “combined structural and dimensional synthesis of task-redundant robot manipulators” the design process of parallel kinematic machines was investigated theoretically and simulatively. In the case of task redundancy, a robot has more movable degrees of freedom than necessary for a task, which is the case, for example, of milling or drilling processes due to the tool rotation. These degrees of freedom are used for optimization in the proposed automated design process. It was shown that this allows the dimensions of the robot structure to be adapted to use this degree of freedom more effectively than in the standard case without optimization. The resulting dimensions can be better than non-optimally dimensioned structures with respect to a target criterion such as maximum actuator force (for actuator selection) or energy consumption. In later applications, energy can thus be saved over the lifetime of the robot, provided that the energy saving potentials created in the design process are also used in the path planning during operation. The developed methods are suitable to automate the design process of parallel kinematic machines and thus to find better and more complete possible realizations of a new robot for a task to be solved in a short time, compared to a manual or only partially automated design process. The software tools developed in the project were published under free license and can be used and further developed by research institutes as well as companies from the fields of special mechanical engineering and automation technology. Suitable fields of application are industrial machining and handling tasks but also niche topics for which there are no suitable commercially available robots. Additional findings of the project concern optimal path planning using both local methods such as gradient projection on the level of acceleration, and global methods such as dynamic programming. The methods are applicable for optimization in the design process as well as in operation.

Publications

 
 

Additional Information

Textvergrößerung und Kontrastanpassung