The engineers at Disney Research developed a computational tool for designing compliant mechanisms that can be 3D produced by 3D printing. The method takes as input a conventional, rigidly-articulated mechanism defining the topology of the compliant design. This input can be both planar or spatial, and supports a number of common joint types which, whenever possible, are automatically replaced with parameterized flexures.
As the technical core of this approach, Disney researchers described a number of objectives that shape the design space in a meaningful way, including trajectory matching, collision avoidance, lateral stability, resilience to failure, and minimizing motor torque. Optimal designs in this space are obtained as solutions to an equilibrium-constrained minimization problem that is solved using a variant of sensitivity analysis.
The researchers demonstrated their method on a set of examples that range from simple four-bar linkages to full-fledged animatronics, and verified the feasibility of the designs by manufacturing (3D printing) physical prototypes. Engineers routinely design for strength and stiffness. Steel and concrete prevent deflections in buildings, and machines resort to rigid articulation in order to avoid deformations.
Although most human designs are inspired by Nature, rigidity is a concept foreign to the living world: from a kangaroo’s legs to the wings of a bat— bones, tendons, and cartilage are the nuts and bolts of organic machines, and deformation is an integral part of the design, crucial for both efficiency and robustness. Unfortunately, designing for flexibility requires deep understanding and precise predictions of finite deformations, which proves to be substantially more difficult than relying on rigidity.
Fueled by progress in technology and computation, however, many fields of engineering have started to embrace deformation and to leverage flexibility for better, more elegant, and ultimately more satisfying designs. Applied to machines, this turn to the flexible leads to compliant mechanisms, i.e., mechanical devices that perform motion not through rigid articulation but by virtue of elastically deforming flexures.
Compliant mechanisms enjoy widespread use in industry, where they are valued for their accuracy, ease of manufacturing, scalability, and cost efficiency. The spectrum ranges from specialized microelectromechanical systems (MEMS) for miniature sensors and actuators, to more mundane devices including monolithic pliers and wiper blades, and to commonplace products such as binder clips, backpack latches, and shampoo lids.
This particular study is primarily interested in exploring the potential of compliant mechanisms for personalized automata and animatronics. With the ability to create complex geometry and its repertoire of flexible, plastic-like materials, 3D printing is an ideal way of manufacturing compliant mechanisms.
Thanks to the increasing availability of consumer-level printers, hobbyist mechanics and other non-expert users now have the machinery to create compliant mechanisms for use in their conceptions and contraptions. But perhaps even more than for conventional mechanisms, the path to a successful compliant design is littered with traps for the novice:
• Compliant mechanisms typically involve large deflections that give rise to nonlinearities in both geometry and material behavior, not rarely betraying intuition.
• For conventional mechanisms, the resistance to motion is either zero or infinite. In the compliant setting, any motion requires a finite amount of work and, depending on direction, the stiffness can vary by orders of magnitude. Shaping the corresponding energy landscape, i.e., finding a balance between stiffness and flexibility is one central aspect of this design problem.
• Compliant mechanisms provide friction- and wear-less motion, but incautious design can induce material fatigue and failure. In order to minimize this risk, high stress concentrations must be avoided.
• While the forward problem of predicting the motion of a compliant mechanism is a non-trivial task already, the inverse problem of determining parameter values that lead to a desired motion or function is extremely difficult.
Considering these challenges, designing compliant mechanisms is all but a hopeless endeavor for casual users