Adapted Pepper

TL;DR

This paper presents Adapted Pepper, a robot prototype with an embedded GPU and additional sensors, enabling advanced perception algorithms to run locally without network reliance, addressing embedded computational power limitations in mass-produced robots.

Contribution

Introduction of a novel Pepper robot prototype with embedded GPU and extra sensors to support real-time, local processing of advanced perception algorithms.

Findings

Successful integration of GPU and sensors into Pepper

Enabling local execution of algorithms like OpenPose and YOLO

Reduced dependency on external computation resources

Abstract

One of the main issue in robotics is the lack of embedded computational power. Recently, state of the art algorithms providing a better understanding of the surroundings (Object detection, skeleton tracking, etc.) are requiring more and more computational power. The lack of embedded computational power is more significant in mass-produced robots because of the difficulties to follow the increasing computational requirements of state of the art algorithms. The integration of an additional GPU allows to overcome this lack of embedded computational power. We introduce in this paper a prototype of Pepper with an embedded GPU, but also with an additional 3D camera on the head of the robot and plugged to the late GPU. This prototype, called Adapted Pepper, was built for the European project called MuMMER (MultiModal Mall Entertainment Robot) in order to embed algorithms like OpenPose, YOLO or to process sensors information and, in all cases, avoid network dependency for deported computation.