This Small Business Innovation Research (SBIR) Phase II project investigates end-user programming for ensembles of robots. The project focuses on the development of an accessible end-user programming environment so that middle and high school students can create their own custom ensembles or blocks of robots and observe how the blocks' behavior affect an entire robot. Building powerful and correct intuitions about the behavior of complex systems is important for scientists and engineers, but with today's technologies it is difficult for children to acquire and integrate these ideas into their mindset. Through exploratory play with the proposed robotics construction kit, which embodies a distributed processing scheme for embedded microprocessors, children can build and observe complex systems acting in the real world. Although end-user programming environments exist for software systems, and even for a few robotics toys, no competing approach to end user programming tackles distributed processing for modular robotics. The project aims to build three experimental systems: a text-based environment, a visual programming language, and a 'cellular automata' interface. Testing with local middle school students will determine the benefits and drawbacks of each approach.
The broader/commercial objective of the project is to give children a vehicle to explore how complex global behaviors emerge from local effects. Designing and building complex systems exposes children to a variety of science, technology, engineering and mathematics (STEM) concepts. The programmed kit, without the end-user programming component proposed here, already introduces these important concepts. The addition of an intuitive, low-threshold, high-ceiling approach to reprogramming individual modules will add extensibility to this already powerful model of complexity. A commercial version of kit will be released in three phases: to science centers and children's museums initially, to a core community of technically savvy enthusiasts, and finally to the public through retail channels. Several science centers have expressed serious and persistent interest in acquiring initial versions of the kits and incorporating them into robotics education programs and exhibits. In addition to the project's primary objective, the design and testing of end-user programming for distributed embedded computing can inform other applications of this technology in the rapidly growing area of modular robotics.
Under this SBIR-II award Modular Robotics LLC developed Cubelets, a kit for building robots that requires no technical skills whatsoever, yet offers beginners an entreé to real robotics and much more. A basic Cubelets kit contains six small blocks that snap together magnetically to make a robot (the master Cubelets kit contains twenty blocks with more functions). Each block has a single function: the kit includes Sense blocks, Action blocks, Think blocks, and Battery block. What is unique about Cubelets among other robot construction kits is how the robot gets its behavior. Unlike other kits, Cubelets robots have no central program, and no single "brain" block. Instead, the behavior of a Cubelets robot depends on how you’ve snapped its various blocks together. Building a robot requires no special mechanical ability, no electronics, and no coding. Yet the simplicity of Cubelets invites playful experimentation with the still-emerging research field of modular robotics, early engagement with engineering, and (we think most fundamentally) powerful ideas about complexity and emergence. The idea that the robot has no brain, yet can interact in complex ways with its environment, suggests an important lesson that transcends specific "STEM" learning goals: Our world in all its complexity arises from many tiny interacting local effects. Outcomes such as global climate change, wildfires, and war are more often the inadvertent results of myriad and seemingly unrelated decisions than the intended outcome of a planned design. Yet because of the ‘top-down’ mindset we’ve been given, we consistently fail to apprehend the ways in which complexity emerges from small and local decisions. The learning goal of Cubelets—beyond the obvious STEM engagement opportunities for people with no technical background—is to foster this understanding of emergent complexity through direct and tactile experience with toys. Modular Robotics is a small independent toy company; Cubelets is our first product to reach the market (others are in development). We’ve had an outstanding market response, and we’re growing our business to meet a burgeoning demand for Cubelets. Yet we want very much to be part of the larger project of preparing young people for the changing world they will face, and we believe that the lessons embedded in Cubelets are essential for understanding this future. The SBIR II project began with a laboratory prototype and finished with a commercial product shipping to customers worldwide. The company redesigned the hardware and software iteratively for manufacturing and to improve quality and reliability of the product; designed and implemented a Web based interface for end-users to reprogram Cubelets using a Bluetooth block to connect personal computers or mobile devices to their Cubelets robots, and tested the hardware and software designs. Several museums and science learning centers have purchased Cubelets and use them in museum activities with children; the Museum of Modern Art (MoMA) included Cubelets in its 2011 exhibit, "Talk to Me". Throughout the project period the company has engaged in numerous outreach and community activities. The company has been commercially successful, beginning the project period with a staff of 2 persons growing to a staff of 16 by the end of the project. The company began taking orders for its first product in June 2011 and by June 2012 orders outstrip manufacturing capacity by a significant margin: the company is expanding its manufacturing capacity to meet demand. The company intentionally followed a bootstrap strategy for the duration of the SBIR project, using NSF funding to grow the value of the company before taking on equity investment.
