Fabrication of Lightweight Link by Modifying Infill Density Based on Topology Optimization

Masajiro Kurihara, Mitsuru Endo, Yukio Tsutsui, Shimpei Tanaka

The 8th Jc-IFToMM International Symposium, The 8th Jc-IFToMM International Symposium

Background

Weight reduction in robots contributes to lower energy consumption and enhances safety by reducing impact forces in the event of a collision. Various approaches have been proposed, such as replacing metal components with lightweight materials or optimizing structure geometries. Among them, topology optimization (TO) is an effective method to generate structures that balance strength and material efficiency.

However, TO outputs often contain intermediate-density regions (grayscale areas) that are difficult to fabricate by traditional manufacturing. Additive Manufacturing (AM) enables these structures to be realized by controlling infill density. In this study, we propose a method for fabricating lightweight structures by directly linking the grayscale density distribution obtained from topology optimization to the infill rate in additive manufacturing. This approach emphasizes manufacturability by constructing a functionally graded structure using variable-density infill corresponding to intermediate density regions.

Method

The topology optimization (TO) was conducted using the SIMP method with sensitivity analysis and the optimality criteria method. The objective was to minimize compliance under a constant volume constraint of 50%. Each finite element’s Each finite element’s density $(\rho)$ was bounded between 0.001 and 1.0 to ensure numerical stability. The penalty parameter $p$ controls the extent of grayscale appearance. Lower values of $p$ allow intermediate densities, which are essential in this study for creating gradations of material distribution that cannot be captured by binary designs.

Fig.1 the design domain and loading conditions.

An optimized structure with $p = 1$ is shown in Fig. 2. Material concentrates near the fixed edge and loaded center.

Fig.2 An example of optimal structure ($p = 1$).

Based on element-wise density, the design domain was classified into five density levels:
(a) $0.01 \leq \rho < \frac{1}{4}$,
(b) $\frac{1}{4} \leq \rho < \frac{1}{2}$, (c) $\frac{1}{2} \leq \rho < \frac{3}{4}$,
(d) $\frac{3}{4} \leq \rho < 1.0$,
(e) $\rho = 1.0$.
Each group was exported as a separate STL model.

Fig.3 Divided models.

Each model corresponding to a different density range was assigned an infill setting in Ultimaker Cura based on its representative average density value. The grid pattern was used with a layer height of 0.4 mm and wall line count set to one. The sliced models were individually printed with PLA filament using FLASHFORGE Adventurer 4/5m Pro. The resulting parts were assembled using a mechanical insert jig. This method enables the reproduction of grayscale density distributions in physical structures through the combination of multiple infill patterns.

Fig.4 Model after setting infill.

The g-code was printed using FLASHFORGE Adventurer 4/5m Pro and PLA filament, resulting in the structure shown below:

Fig.5 Cross section of printed structure.

Results

Compression tests were conducted using a 6-ton press and load cell. Structures with varying penalty parameters ($p = 0.5, 1.0, 3.0$) and a reference structure without TO were fabricated and evaluated.

The beam with ($p = 3$) showed the highest strength despite its lower compliance performance in simulation. Fig. 9 shows the fractured cross-section for (p = 0.5), where porous zones deformed under load. This discrepancy is attributed to three factors:

Boundary mismatch: Simulated fixed support differs from localized experimental contact, reducing load dispersion.
Infill pattern limitations: The grid pattern does not faithfully replicate grayscale transitions.
Print direction: Layers were perpendicular to the compression direction, causing premature buckling in porous areas.

Conclusion

This study demonstrated a method to fabricate lightweight robotic structures by transferring grayscale outputs of TO into infill densities for AM. While the compliance improved in low-penalty models, mechanical testing showed that only denser structures achieved high strength. This is due to boundary condition effects, infill fidelity, and print orientation.

Further refinements—such as better modeling of boundary conditions, directional printing, and alternative infill patterns—will be needed to fully exploit the potential of grayscale structures.

References

[1] Antonio Bicchi et al.: Fast and “Soft-Arm” Tactics, IEEE Robotics & Automation Magazine, 2004, pp. 22-33.
[2] Thomas Lens et al., Investigation of Safety in Human-Robot-Interaction for a Series Elastic, Tendon-Driven Robot Arm, IROS, 2012, pp. 1-6.
[3] Otani Takuya et al., Energy efficiency improvement of a robotic finger with ultra high molecular weight polyethylene gear. IEEE Access, 2022, pp. 100033–100040.
[4] Haibin Yin et al., Hybrid Structure Design of Lightweight Robotic Arms Based on Carbon Fiber Reinforced Plastic and Aluminum Alloy, IEEE Access, 2019, pp. 64932–64945.
[5] Mustafa Bugday et al., Design optimization of industrial robot arm to minimize redundant weight, Engineering Science and Technology, an International Journal, 2019, pp. 346–352.
[6] Jihong Zhu et al., A review of topology optimization for additive manufacturing: Status and challenges, CSAA, 2021, pp. 91–110.
[7] Matthijs Langelaar, Topology optimization of 3D self-supporting structures for additive manufacturing, Additive Manufacturing 12, 2016, pp. 60–70.
[8] Nishiwaki Shinji et al., Topology Optimization, Maruzen Publishing, 2013, p. 46, pp. 59–86.
[9] Bao Yuanhao et al., Development of a combination beam based on a new combined method of different filling densities of 3D printing, Robotics and Mechatronics, 2024.
[10] O. Sigmund, A 99 line topology optimization code written in Matlab Structural and Multidisciplinary Optimization, Struct Multidisc Optim 21, 2001, pp. 120–127.

The content on this page is not freely available for reproduction or redistribution. Unauthorized use may lead to legal consequences.