A Hetero-Associative Sequential Memory Model Utilizing Neuromorphic Signals: Validated on a Mobile Manipulator

TL;DR

This paper introduces a neuromorphic hetero-associative memory system for mobile manipulators that efficiently learns tactile and joint state associations, enabling adaptive, low-cost control and grasping behaviors demonstrated on a real robot.

Contribution

It presents a novel neuromorphic encoding and associative memory approach with 3D rotary positional embeddings for tactile-based robot control and grasp sequence retrieval.

Findings

Successfully controls robot movement based on tactile input

Retrieves multi-joint grasp sequences from tactile data

Demonstrates generalization and quick setup on real robot

Abstract

This paper presents a hetero-associative sequential memory system for mobile manipulators that learns compact, neuromorphic bindings between robot joint states and tactile observations to produce step-wise action decisions with low compute and memory cost. The method encodes joint angles via population place coding and converts skin-measured forces into spike-rate features using an Izhikevich neuron model; both signals are transformed into bipolar binary vectors and bound element-wise to create associations stored in a large-capacity sequential memory. To improve separability in binary space and inject geometry from touch, we introduce 3D rotary positional embeddings that rotate subspaces as a function of sensed force direction, enabling fuzzy retrieval through a softmax weighted recall over temporally shifted action patterns. On a Toyota Human Support Robot covered by robot skin, the hetero-associative sequential memory system realizes a pseudocompliance controller that moves the link under touch in the direction and with speed correlating to the amplitude of applied force, and it retrieves multi-joint grasp sequences by continuing tactile input. The system sets up quickly, trains from synchronized streams of states and observations, and exhibits a degree of generalization while remaining economical. Results demonstrate single-joint and full-arm behaviors executed via associative recall, and suggest extensions to imitation learning, motion planning, and multi-modal integration.