Source-linked AI summary
TO-mdiSPAs: Topology Optimization of multi-directional Soft Pneumatic Actuators
Swagatam Islam Sarkar, Prabhat Kumar
TL;DR
The paper addresses the limited availability of systematic SPA design methods for multidirectional motion under design-dependent pneumatic loading. It uses robust topology optimization with Darcy-based pressure modeling, and simulations show an unconventional optimized design achieving multidirectional bending and gripper motion.
Problem
Systematic SPA design methods are limited, while many existing actuators are restricted to planar motion and pneumatic loads change with the design.
Source: §I · pp. 1–2Method
The paper uses robust topology optimization with Darcy’s law and a drainage term to model design-dependent pneumatic loads.
Source: §I · p. 2, §II · p. 2, §III · p. 2Results
The optimized chamber geometry differs from a conventional rectangular configuration, and simulations show multidirectional bending and a gripper capable of adaptive, compliant grasping.
Source: §IV · pp. 4, 6, §V · p. 6Takeaways & Limitations
The optimized mdiSPA demonstrates a topology-optimized design for multidirectional pneumatic actuation, including use as a gripper mechanism.
Source: §V · p. 6
Abstract
from arXiv · showhide
Soft pneumatic actuators (SPAs) are highly promising and have been extensively explored within the field of soft robotics. Under pneumatic pressure loads, an SPA undergoes bending deformation to perform specific tasks. A multi- or omni-directional SPA can bend and move in any direction within a 3D space, leveraging its multiple degrees of freedom to realize various mechanical functions. This work presents a systematic methodology using topology optimization (TO) to achieve an optimized design for a multi-directional SPAs. To ensure manufacturing robustness, a three-field TO formulation considering blueprint, dilated, and eroded designs is implemented. Additionally, Darcy's law, incorporating a drainage term, is used to model the design-dependent nature of the pneumatic loading. A min-max optimization problem is formulated based on the target output deformations of the SPA unit and solved using the Method of Moving Asymptotes. The optimization yields a high-performing, unconventional geometric design. Finally, numerical simulations demonstrate that the optimized SPA successfully achieves versatile multi-directional movements.
I. INTRODUCTION
Soft pneumatic actuators offer flexible motion, but many designs rely on heuristic principles and remain limited to planar motion. This work applies topology optimization to systematically design multi-directional SPAs while accounting for design-dependent pneumatic loads.
I. INTRODUCTION
- Soft pneumatic actuators are pressure-driven soft robots, one of several actuation classes that also include shape-memory, cable-driven, and dielectric-elastomer actuation.SPAs are classified by motion as extending or contracting, bending, helical or twisting, and bidirectional or omnidirectional.Source: §I · p. 1
- Prior work includes multi-directional actuators and soft actuators capable of bending in multiple directions, but the cited designs are primarily heuristic.Examples include actuators with omnidirectional bending and a three-chamber design combining extension and omnidirectional motion.Source: §I · p. 1
- The work introduces topology optimization for systematic design of multi-directional SPAs, addressing reliance on heuristic designs and limitations to planar motion.The framework accounts explicitly for design-dependent pneumatic actuation.Source: §I · p. 1
- Pneumatic loads change location, direction, and magnitude as the design evolves, creating a challenge for topology optimization.The design domain assigns each element a variable ρ, where 1 is solid and 0 is void.Source: §I · p. 2
- The paper builds on TO methods for pressure-loaded compliant mechanisms and extends prior bi-directional SPA actuation toward motion in arbitrary directions.It uses a Darcy-based method and a robust formulation intended to address manufacturing tolerances.Source: §I · pp. 1–2
II. MODELING OF PRESSURE-INDUCED PNEUMATIC
The pressure model uses Darcy’s law and a drainage term to represent pneumatic loading as the design evolves. The resulting pressure field is converted into nodal forces that drive structural deformation.
II. MODELING OF PRESSURE-INDUCED PNEUMATIC
- Darcy’s law models pneumatic loading across evolving solid and void regions in the finite-element design.The flux depends on permeability, fluid viscosity, and the pressure gradient.Source: §II · p. 2
- A drainage term induces localized pressure gradients near solid–void interfaces.The term depends on the pressure difference between instantaneous and external pressure and on a design-dependent drainage coefficient.Source: §II · p. 2
- Finite-element assembly forms a global flow system whose pressure field yields nodal forces that drive structural deformation.Those forces are related to structural displacement through the global stiffness and transformation matrices.Source: §II · p. 2
III. TOPOLOGY OPTIMIZATION OF SPA
The density-based topology optimization framework interpolates material stiffness and uses a Heaviside projection filter to obtain the physical design variable. The illustrated domains include a symmetric quarter-arm and a symmetric half-chamber.
III. TOPOLOGY OPTIMIZATION OF SPA
- Young’s modulus is interpolated between solid and void material using the modified SIMP approach.The void modulus is kept very small to avoid singularity in the stiffness matrix.Source: §III · p. 2
- A Heaviside projection filter maps the filtered design variable to the physical design variable for each element.The physical design variable is denoted by ρ̄e and the filtered variable by ρ̃e.Source: §III · p. 2
- The optimization domains shown are a symmetric quarter of an arm with multiple chambers and a symmetric half of one chamber.Source: Figure 1 · p. 3, Figure 2 · p. 3
A. multidirectional SPA
The design uses a symmetric quarter-arm domain and a symmetric half-section of one pressure-loaded chamber. Four directional channels can be activated separately or together to produce multidirectional movement.
A. multidirectional SPA
- A symmetric quarter-arm design domain represents an arm with multiple chambers, with non-design material at the bottom and arrows indicating desired output and pressure-load directions.The actual arm uses ten chambers.Source: A. multidirectional SPA · p. 3
- A symmetric half-section of one chamber is pressure-loaded, with output displacement measured at a location against a linear spring representing the interacting workpiece.The lower-left and right-edge regions are non-design domains.Source: A. multidirectional SPA · p. 3
- Four channels correspond to the front, back, left, and right directions; activating one channel moves the arm opposite that channel.Source: A. multidirectional SPA · p. 3
- Activating multiple channels with equal or different pressures combines directional movements to produce output in other directions.Equal pressure in the front and right channels produces a 45° trajectory toward the back-left.Source: A. multidirectional SPA · p. 3
B. TO formulation
The robust topology optimization formulation minimizes the worst output displacement across blueprint and eroded designs while enforcing design constraints. The design variables are updated using MMA with adjoint-based sensitivities.
B. TO formulation
- The robust formulation accounts for blueprint and eroded configurations, with the displacement evaluated using a virtual load applied at the output degree of freedom.The virtual load vector has all entries zero except the output degree of freedom, set to 1.Source: B. TO formulation · p. 3
- The objective minimizes the maximum output displacement across the blueprint and eroded designs.The two terms are LT Ub and LT Ue, respectively.Source: B. TO formulation · p. 3
- The formulation constrains blueprint-design volume and eroded-design strain energy against their respective allowable limits.The associated Lagrange multipliers correspond to these constraints.Source: B. TO formulation · p. 3
- The Method of Moving Asymptotes updates design variables using sensitivities computed with the adjoint-variable method.Source: B. TO formulation · p. 3
- The figure captions identify an optimized single chamber under pressure and a 3D arm composed of ten chambers in each of four directions.Source: Figure 3 · p. 4, Figure 4 · p. 4
IV. RESULTS AND DISCUSSIONS
The optimized chamber geometry differs from the conventional rectangular design, and simulations show that the three-arm gripper bends in multiple directions under equal and unequal channel pressures.
IV. RESULTS AND DISCUSSIONS
- The optimized chamber geometry notably differs from the conventional rectangular configuration.The optimization used a blueprint volume-fraction limit of 0.20 and constrained the eroded design’s strain-energy fraction to 80%.Source: §IV · p. 4
- The optimized SPA is represented as a gripper with three interconnected arms, and its multiple arms support adaptive, compliant grasping.The design was converted from the optimized topology into a 2D CAD model and then extruded into 3D chamber units for the arms.Source: §IV · pp. 4, 6, Figure 5 · p. 4
- Equal-pressure and unequal-pressure simulations show that each arm’s tail bends near the resultant direction of its applied pressure, supporting multidirectional gripping.In the equal-pressure case, arm 2 bends mainly opposite its pressurized backside chambers; the unequal-pressure case likewise shows tail orientations near the stated resultant pressure directions.Source: §IV · pp. 4–5, Figure 6 · p. 5, Figure 7 · p. 5
V. CONCLUSIONS
The work presents a topology-optimized multidirectional soft pneumatic actuator and develops it into a gripper that moves and grips in multiple directions.
V. CONCLUSIONS
- The topology-optimized mdiSPA is developed into a gripper mechanism that can move and grip in multiple directions.The design maximizes output displacement while constraining strain energy in the eroded design and limiting blueprint material volume; Darcy’s law models design-dependent pneumatic loading.Source: §V · p. 6
- The developed mdiSPA achieves multidirectional bending under pneumatic actuation, with additive-manufacturing fabrication and experimental validation identified as future work.The conclusion presents gripper use as an application of the actuator’s multidirectional bending.Source: §V · p. 6