2022 Fiscal Year Final Research Report
Estimation of soil water content using LiDAR intensity
Project/Area Number |
21K17838
|
Research Category |
Grant-in-Aid for Early-Career Scientists
|
Allocation Type | Multi-year Fund |
Review Section |
Basic Section 61050:Intelligent robotics-related
|
Research Institution | 防衛大学校(総合教育学群、人文社会科学群、応用科学群、電気情報学群及びシステム工学群) |
Principal Investigator |
Eto Ryosuke 防衛大学校(総合教育学群、人文社会科学群、応用科学群、電気情報学群及びシステム工学群), システム工学群, 講師 (20761480)
|
Project Period (FY) |
2021-04-01 – 2023-03-31
|
Keywords | LiDAR / 含水比 / ガウス過程回帰 |
Outline of Final Research Achievements |
The purpose of this study is to develop a method to estimate the water content ratio of soil surface using LiDAR intensity data. First, I measured the relationship between the LiDAR laser irradiation angle, water content, and LiDAR intensity data for Toyoura sand, volcanic ash, and Masa soil in laboratory experiments. Water content estimation models were generated from the obtained data using Gaussian process regression. In laboratory experiments, the estimated value by the model was found to be within ±5 points of error in 70% to 98% of the cases. In outdoor experiments with a robot equipped with a LiDAR, the model was able to estimate the water content within ±2 to 4 points of error.
|
Free Research Field |
ロボット工学
|
Academic Significance and Societal Importance of the Research Achievements |
非接触での含水量推定は2波長の近赤外光の受光強度を用いて推定されている.しかし,使用される波長は1500 nm付近や2000 nm付近のものであり高価なセンサが必要であった.本研究の提案手法は,近年無人ロボットに搭載されているLiDARの受光強度のみを用いて土壌含水量を推定するものである.そのため,ロボットの周辺土壌の含水量を推定するために新たに高価なセンサを必要とせず,ロボットの簡素化,低廉化に貢献できる.
|