Neuromorphic circuit for temporal odor encoding in turbulent environments

TL;DR

The paper tackles robust real-time odor concentration sensing in turbulent plumes by analyzing MOx sensor responses to artificial plumes and proposing a neuromorphic front-end that encodes concentration as spike-timing differences between two parallel pathways, inspired by the mammalian olfactory bulb. It introduces a bout-slope feature, derived from band-pass filtered MOx signals, that remains informative across plume dynamics and tests a modified circuit design that extracts and encodes these slopes into discrete spikes. Results show that bout slopes correlate with concentration in single-pulse data and that multi-sensor, combinatorial outputs improve gas recognition in plume environments, albeit with limitations near peak overlap and history effects. The work advances data-efficient, real-time e-nose capabilities for tasks like environmental monitoring and search-and-rescue, while outlining pathways to more metabolically plausible spiking models for flexible plume navigation.

Abstract

Natural odor environments present turbulent and dynamic conditions, causing chemical signals to fluctuate in space, time, and intensity. While many species have evolved highly adaptive behavioral responses to such variability, the emerging field of neuromorphic olfaction continues to grapple with the challenge of efficiently sampling and identifying odors in real-time. In this work, we investigate Metal-Oxide (MOx) gas sensor recordings of constant airflow-embedded artificial odor plumes. We discover a data feature that is representative of the presented odor stimulus at a certain concentration - irrespective of temporal variations caused by the plume dynamics. Further, we design a neuromorphic electronic nose front-end circuit for extracting and encoding this feature into analog spikes for gas detection and concentration estimation. The design is inspired by the spiking output of parallel neural pathways in the mammalian olfactory bulb. We test the circuit for gas recognition and concentration estimation in artificial environments, where either single gas pulses or pre-recorded odor plumes were deployed in a constant flow of air. For both environments, our results indicate that the gas concentration is encoded in -- and inversely proportional to the time difference of analog spikes emerging out of two parallel pathways, similar to the spiking output of a mammalian olfactory bulb. The resulting neuromorphic nose could enable data-efficient, real-time robotic plume navigation systems, advancing the capabilities of odor source localization in applications such as environmental monitoring and search-and-rescue.

Figures (15)

• Figure 1: PID traces of plumes used to control odor valves of gases G1 and G2. The y-axis depicts the gas concentration level measured by PID relative to the highest used gas concentration. The number associated with the label of each PID trace represents the trial number.
• Figure 2: MOx recordings for plume traces 1 and 2 for different gas combinations.
• Figure 3: Band-pass filtered MOx recordings for the plume trace Eu_ EB_ 2. The largest bouts in the duration 0-2s and 2-5s are highlighted in red.
• Figure 4: Bar plots of bout slopes for all gases at two different peaks in the plume traces shown in Table 1.
• Figure 5: Scatter plot of bout slopes for all gases at two different concentration levels at Peak 1. The x-axes represent different concentration levels in the form of percentages relative to the highest used gas concentration. The interpolated red dotted line depicts the rising trend for bout slopes of Eu gas with respect to gas concentration level.