Reverse Engineering the Octopus Arm

Abstract

Despite the extreme flexibility of the octopus’s arms and their resulting near infinite possible configurations, the octopus effectively controls its arms for use in a wide range of behaviors, including locomotion, foraging, excavation, exploration, and manipulation. These behaviors also gain a number of advantages from this extreme mechanical flexibility. The octopus not only successfully controls limbs with infinite degrees of freedom, but prevalently exploits this biomechanical property to generate adaptive behavior. The octopus therefore serves as an ideal model for soft robotics. If appropriately characterized, the octopus’s biomechanical properties and control strategies could be implemented in the development of a soft robotic limb with the same range of capabilities. Many of the octopus’s behaviors, such as when foraging at night or in visually occluded spaces, are executed under conditions of limited or absent visual feedback. In such conditions, the octopus must rely primarily on the complex chemotactile system of its suckers to successfully find and capture prey. The investigations presented here aimed to characterize the search strategies used by the octopus in such conditions and identify the primary control mechanisms driving these strategies. Sucker recruitment was revealed to be a key mechanism employed during search, and one that can be adapted to perform systematic search patterns over complex surfaces. This mechanism also provides a number of other behavioral and computational advantages to the octopus, which are discussed, and was therefore implemented in the development of a soft robotic arm. Inspired by the repeated sucker units of the octopus arm, this robotic arm was created by designing the mechanical properties and control algorithm of a single segment, then simply replicating this segment and assembling the replicates end-to-end.