Perceive, Represent, Generate: Translating Multimodal Information to Robotic Motion Trajectories

TL;DR

This paper introduces PRG, a three-stage framework that translates multimodal perceptual inputs into robotic motion trajectories, demonstrated through a handwriting task with inputs like images and sounds.

Contribution

The paper presents a novel three-stage framework combining perception, multimodal encoding, and trajectory generation for robotic motion based on multimodal instructions.

Findings

Effective translation of multimodal inputs into handwriting trajectories

Successful implementation on a robotic platform for writing tasks

Demonstrated coherence and readability in generated handwriting

Abstract

We present Perceive-Represent-Generate (PRG), a novel three-stage framework that maps perceptual information of different modalities (e.g., visual or sound), corresponding to a sequence of instructions, to an adequate sequence of movements to be executed by a robot. In the first stage, we perceive and pre-process the given inputs, isolating individual commands from the complete instruction provided by a human user. In the second stage we encode the individual commands into a multimodal latent space, employing a deep generative model. Finally, in the third stage we convert the multimodal latent values into individual trajectories and combine them into a single dynamic movement primitive, allowing its execution in a robotic platform. We evaluate our pipeline in the context of a novel robotic handwriting task, where the robot receives as input a word through different perceptual modalities (e.g., image, sound), and generates the corresponding motion trajectory to write it, creating coherent and readable handwritten words.