Canidae Mechanica, a project that began a few months after the development of Paradigm Black, is the second generation robotic limb prototype created to build upon the knowledge and philosophy I initially encountered with my first foray into robotics.
Much larger than the previous 21 centimeter design, this new machine stands at 58 centimeters tall, and involves over 80 individual parts. Designed after the anatomy of a canine forelimb, this iteration expands upon its predecessor by being constructed of many slim and redundant struts that replicate properties found in natural bone structures.
Each individual strut is made of laser cut Delrin acetal homopolymer resin, a material which, unlike many plastics is able to bend and hold up under stresses extremely well. Layering struts of varying thicknesses, lengths and curvatures creates a structure that can absorb repeated shocks while still remaining strong and lightweight - properties commonly found in organic skeletal structures.
One of the design's most prominent features is its rather large assortment of claws located at the foot of the limb. Each claw is connected to a series of struts that tie them into the overall leg structure, and as the limb is contracted upwards the claws will curl inward, replicating the actual movement found in many canines and felines. When on the ground, their pointed tips easily sink into soft terrain, adding crucial traction needed when pushing forward into a walk cycle.
When it came time to physically construct Canidae Mechanica, I started by assembling the electronic circuits that would later be programmed to control the limb's horizontal and vertical movements. Rather than implementing imprecise DC motors for this task, I instead decided to try using stepper motors - relatively new forms of electronic motors that are currently used mainly in 3D printers. Each motor is connected to one Arduino processing board and one stepper driver board. In an effort to make the circuitry fit into a more compact space, the boards have been oriented into a cube-like assembly. I call this the Motion Matrix Unit - or MMU.
Unfortunately, when it came time to test the motion of the prototype, I found that the stepper motors that I had chosen for the job were just barely able to lift the weight of the limb during its vertical movement. The horizontal motion was successful, however it seems that the supplier of the two metal air pistons used as shock absorbers for the leg accidentally listed them as being made of aluminum, not steel. Thus several unaccounted pounds were added to the system. Another prime example of the many difficulties one will encounter when prototyping.
Despite the technical difficulties experienced, the fully constructed machine was displayed in Purdue's Rueff Gallery as part of their 2016 Introspectives exhibition. Displayed along with several of my other works now archived on this site, Canidae Mechanica attempts to not only explore the relationship between robotics and nature, but the place of robotics in the traditional art gallery.
