Vanteddu, P. R., Nava, G., Bergonti, F., L’Erario, G., Paolino, A., & Pucci, D. (2024). From CAD to URDF: Co-Design of a Jet-Powered Humanoid Robot Including CAD Geometry. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
This research paper aims to address the limitations of traditional co-design optimization strategies in robotics, which often rely on simplified robot models and may result in suboptimal mechanical designs. The authors propose a framework that integrates CAD and URDF models to optimize the design of a jet-powered humanoid robot for improved flight performance while ensuring structural safety.
The researchers developed a co-design framework that utilizes a multi-objective evolutionary algorithm (NSGA-II) to optimize the geometric parameters of critical CAD components of the iRonCub robot, specifically those related to jet interfaces. The framework incorporates an automated Finite Element Method (FEM) analysis to filter out designs that do not meet the required structural safety margins. The optimization process evaluates the performance of each design candidate using flight simulations based on a previously optimized flight controller.
The proposed co-design framework successfully generated optimized designs that outperformed the original design in terms of momentum tracking, overall thrust consumption, and velocity error. The integration of FEM analysis ensured the structural integrity of the optimized designs. The researchers demonstrated the effectiveness of their framework by validating the optimized designs using various flight trajectories.
The study highlights the importance of integrating detailed CAD geometry and structural analysis into the co-design optimization process for robots, especially for complex systems like flying humanoids. The proposed framework provides a systematic approach to optimize both the control performance and mechanical design of robots, leading to improved overall performance and safety.
This research contributes to the field of robotics by presenting a practical and effective co-design framework that bridges the gap between simplified models and real-world prototyping. The findings have implications for the design and development of future robots, particularly those operating in challenging environments where both control and mechanical design are critical for success.
The current framework focuses on optimizing specific components of the robot. Future research could explore extending the optimization to encompass more robot parts and incorporate additional design parameters. Further investigation into dynamic FEM analysis and the inclusion of control parameters in the optimization process could lead to even more robust and high-performing robot designs.
To Another Language
from source content
arxiv.org
Key Insights Distilled From
by Punith Reddy... at arxiv.org 10-11-2024
https://arxiv.org/pdf/2410.07963.pdfDeeper Inquiries