LiDAR Data Evaluation Model for Carbon Storage in Artificial Forest Based on Reinforcement Learning

View at publisher Summarise with VIRAT
Authors: Beilei Yin, Hazandy Abdul Hamid, Norizah Kamarudin, Li Liangcai, Liu Feng, Liao Hailing
Cited by: 0
References: 2

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper proposes a LiDAR data evaluation model for carbon storage in artificial forest based on reinforcement learning, and designs an innovative reinforcement learning algorithm. In the algorithm, the state space and action space are redefined to fit the characteristics of LiDAR data, a reward function containing multiple indicators is constructed to guide the learning of the agent, and the adaptive learning rate adjustment and experience replay improvement strategy are used to optimize the algorithm performance. In the experimental stage, a LiDAR data set of artificial forest covering different regions and forest ages was constructed, and the algorithm in this paper was compared with the traditional regression model and the existing reinforcement learning related algorithms. The experimental results show that in terms of mean square error index, the algorithm in this paper is 42.3% lower than the traditional multivariate linear regression model and 28.7% lower than the carbon storage evaluation algorithm based on Q-learning, the mean absolute errors are reduced by 39.6% and 25.4% respectively, the determination coefficient reaches 0.921, which is significantly improved compared with other algorithms, verifying the advantage of the algorithm in the accuracy of carbon storage evaluation. At the same time, the convergence speed of the algorithm is about 35% higher than that of the comparison algorithm, and it performs stably on different data sets and has strong generalization ability.

Fields of Study & Subjects

Remote Sensing and LiDAR Applications Forest ecology and management Forest Management and Policy

Publication details

Journal / source
IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
Pages
1642–1647
Publication date
28 January 2026
Language
EN