Abstract
This paper presents a neuromorphic tactile encoding methodology that utilizes a temporally precise event-based representation of sensory signals. We introduce a novel concept where touch signals are characterized as patterns of millisecond precise binary events to denote pressure changes. This approach is amenable to a sparse signal representation and enables the extraction of relevant features from thousands of sensing elements with sub-millisecond temporal precision. We also proposed measures adopted from computational neuroscience to study the information content within the spiking representations of artificial tactile signals. Implemented on a state-of-the-art 4096 element tactile sensor array with 5.2 kHz sampling frequency, we demonstrate the classification of transient impact events while utilizing 20 times less communication bandwidth compared to frame based representations. Spiking sensor responses to a large library of contact conditions were also synthesized using finite element simulations, illustrating an 8-fold improvement in information content and a 4-fold reduction in classification latency when millisecond-precise temporal structures are available. Our research represents a significant advance, demonstrating that a neuromorphic spatiotemporal representation of touch is well suited to rapid identification of critical contact events, making it suitable for dynamic tactile sensing in robotic and prosthetic applications.
Highlights
The increased expectation of robots to be able to interact in natural dynamic environments requires the ability to quickly detect, recognize and respond to physical contact events
We present our findings from the FEM simulations and physical experiments, with an emphasis on how information may be lost due to reduced temporal resolution
This difference in performance indicates that while a sampling period of 20 ms is within Nyquist limits for detecting the indentation event, the spatiotemporal features needed for fine discrimination is contained at time scales notably shorter than the duration of a contact event
Summary
The increased expectation of robots to be able to interact in natural dynamic environments requires the ability to quickly detect, recognize and respond to physical contact events. Such capabilities are necessary for dexterous manipulation (Johansson, 2002) and to ensure the safety of the user and robot (Lumelsky, 2005; Dahiya et al, 2013). The entire exterior of the robot should be sensitive and responsive to touch These requirements highlight the need for a rapid acquisition and processing mechanism to interface with thousands of tactile sensing elements distributed across the surface of a robot. A practical approach is to intelligently reduce the data generated from the sensing front-end, prior to transmission
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have