First Author:Haibo Sun

Correspondence Author:Feng Zhu

Co author:YangyangLi,PengfeiZhao,Yanzi Kong,Jianyu Wang,Yingcai Wan,Shuangfei Fu

Journal:Frontiers in Neurorobotics

Volume:17

DOI number:10.3389/fnbot.2023.1093132

Affiliation of Author(s):Faculty of Robot Science and Engineering, Northeastern University

Teaching and Research Group:物理化学

Place of Publication:SWITZERLAND

Abstract:Active object recognition (AOR) provides a paradigm where an agent can capture additional evidence by purposefully changing its viewpoint to improve the quality of recognition. OneofthemostconcernedproblemsinAORisviewpointplanning (VP) which refers to developing a policy to determine the next viewpoints of the agent. A research trend is to solve the VP problem with reinforcement learning, namely to use the viewpoint transitions explored by the agent to train the VP policy. However, most research discards the trained transitions, which may lead to an ine cient use of the explored transitions. To solve this challenge, we present a novel VP method with transition management based on reinforcement learning, which can reuse the explored viewpoint transitions. To be specific, a learning framework of the VP policy is first established via the deterministic policy gradient theory, which provides an opportunity to reuse the explored transitions. Then, we design a scheme of viewpoint transition management that can store the explored transitions and decide which transitions are used for the policy learning. Finally, within the framework, we develop an algorithm based on twin delayed deep deterministic policy gradient and the designed scheme to train the VP policy. Experiments on the public and challenging dataset GERMS show thee ectiveness of our method in comparison with several competing approaches.

Key Words:active object recognition, viewpoint planning, deterministic policy gradient, twin delayed deep deterministic policy gradient, viewpoint transition management, reinforcement learning

Document Code:WOS:000950219200001

Discipline:Natural Science

First-Level Discipline:Chemistry

Page Number:1093132

ISSN No.:1662-5218

Translation or Not:no