Multi-style Learning for Adaptation of Perception Intelligence in Home Service Robots

Qi Wang,Senlin Zhang,Weihua Sheng,Badong Chen,Meiqin Liu

DOI: https://doi.org/10.1016/j.patrec.2021.08.026

IF: 4.757

2021-01-01

Pattern Recognition Letters

Abstract:Robots need more intelligence to complete perception tasks in uncertain and unstructured environments. This paper presents a new self-evolving home service robot framework that learns new perception skills by using manually labeled data obtained in a new home environment. In this framework, a global model is trained which serves as the starting point of the robots' local model, and an adaptation mechanism is developed in the robot to adapt the initial local model to the new home environment. First, three different data sampling styles are proposed to carry out the adaptation process, and theoretical analysis is given to explain the difference between the proposed three data sampling styles. Second, the most suitable data sampling style for incremental learning for a home service robot is determined. Third, we present a case study of multi-style learning, and the experimental results validate our analysis. The theoretical analysis and experimental results lead us to propose a guideline of data collection and labeling for robots to adapt their perception intelligence in home environments. (c) 2021 Elsevier B.V. All rights reserved.