Build-a-Bot: Computer-Aided Interactive Robot Design
We aim to democratize robotics by making engineering design intuitive and accessible to everyone through computation. Currently robot design is a time and skill intensive process requiring years of training and multiple iterations of prototyping and testing. Unlocking the full potential of ubiquitous robotics requires new intelligent design tools that support experienced and novice users alike in the design process, from ideation to prototyping to validation stages.
Using a combination of computational geometry, mechanics, and data compression techniques, we create computational frameworks for designing, fabricating, and controlling legged robots that are robust to task, environment, and fabrication uncertainties. Unlike static objects, legged robots rely on complex interactions with their surrounding environment to locomote, and their performance is therefore very dependent on their physical designs. We focus on algorithms for design synthesis that provide explainable designs based on evaluated efficiency and robustness tradeoffs.
Geometric Modeling and Optimization of Compliant Mechanisms
We leverage the inherent mechanics of robots to make them robust and energy-efficient. Compared to traditional rigid robot designs, compliant robots offer greater flexibility and adaptability, but this comes at a high cost of design and fabrication complexity. Our strategy is to design new compliant mechanisms whose inherent mechanics are controlled by geometric parameters, enabling engineers to tune a robot’s natural response independently of its materials and method of fabrication, or a robot to tune its own behavior on site.
We are particularly interested in kirigami tessellations and lattices, which are able to change their bulk stiffness by orders of magnitude. We develop reduced-parameter models for these designs and demonstrate both theoretically and experimentally how these parameters can be optimized for specific applications such as manipulation, jumping, and impact absorption.
Rapid Fabrication and Assembly of Folding Machines
We investigate new ways to use and combine existing digital manufacturing tools such as 3D printing, lamination, and origami-inspired assembly to make robots within hours. Print-and-fold manufacturing promises inexpensive and customizable robots with embedded actuation, sensing, and electronics. We create new systems for streamlining this process by embedding fabrication constraints into computational tools, automatically converting designs into valid fabrication plans that can be sent straight to the printer. By taking advantage of active materials such as shape memory polymers, we also develop new self-folding and self-assembly processes to turn virtual robot designs into physical designs with minimal human assembly.