Meeting Abstract
P2.28 Wednesday, Jan. 5 A Tunable, Multisegmented Robotic Antenna for Identifying and Testing Biomechanical Design Principles DEMIR, A.*; SAMSON, E.G.; MONGEAU, J.-M.; JAYARAM, K.; FULL, R.J.; COWAN, N.J.; Johns Hopkins University; Johns Hopkins University; University of California, Berkeley; University of California, Berkeley; University of California, Berkeley; Johns Hopkins University ncowan@jhu.edu
Arthropod antennae are complex sensory structures whose mechanical properties are well tuned for certain behavioral tasks. Cockroaches follow walls at speeds up to 65 cm/s, executing up to 25 turns/s, exemplifying this task-level biomechanical tuning. We developed a robotic research platform to explore how mechanical parameters of an antenna may affect task-level performance such as wall following and surface contour extraction. The robotic antenna model is multi-segmented, modular, and can detect its own bending and external contact. Many parameters can be altered: axial and radial stiffness profile, mass distribution, length, and material. We examined how the proximal-distal antenna stiffness distribution affects spatial acuity, and how it affects the correlation of the point of maximum bend with the point of contact. We also tested how readily antennae with different stiffness profiles adopt the shape characteristic of cockroach wall following. We configured the robotic antenna with three distinct stiffness distributions: uniform, linearly decreasing, and logarithmically decreasing. We found that an antenna with logarithmically decreasing stiffness exhibits the best performance (with respect to acuity, bending position, and shape transition), followed by one with linearly decreasing stiffness. This work focuses on the mechanical design features that enable biological hypothesis generation; our results are reported with companion biological experiments in Mongeau et al. (SICB 2011). This versatile platform enables us to identify and isolate the dominant mechanical properties that contribute to biological antennal behavior and pave the way for the creation of robust robotic antennae that are mechanically tuned for user-specified tasks.
