Abstract
Energy-efficient simultaneous localization and mapping (SLAM) is crucial formobile robots exploring unknown environments. The mammalian brain solves SLAMvia a network of specialized neurons, exhibiting asynchronous computations andevent-based communications, with very low energy consumption. We propose abrain-inspired spiking neural network (SNN) architecture that solves theunidimensional SLAM by introducing spike-based reference frame transformation,visual likelihood computation, and Bayesian inference. We integrated ourneuromorphic algorithm to Intel's Loihi neuromorphic processor, a non-VonNeumann hardware that mimics the brain's computing paradigms. We performedcomparative analyses for accuracy and energy-efficiency between ourneuromorphic approach and the GMapping algorithm, which is widely used in smallenvironments. Our Loihi-based SNN architecture consumes 100 times less energythan GMapping run on a CPU while having comparable accuracy in head directionlocalization and map-generation. These results pave the way for scaling ourapproach towards active-SLAM alternative solutions for Loihi-controlledautonomous robots.