Hu, Difeng;
Gan, Vincent JL;
Wang, Tao;
Ma, Ling;
(2022)
Multi-agent robotic system (MARS) for UAV-UGV path planning and automatic sensory data collection in cluttered environments.
Building and Environment
, 221
, Article 109349. 10.1016/j.buildenv.2022.109349.
Preview |
Text
Manuscript_LM.pdf - Accepted Version Download (4MB) | Preview |
Abstract
There has been growing interest in increasing the application of robotic and automation technologies for building indoor inspection. However, much previous research on indoor robotic applications was limited to a single type of unmanned aerial/ground vehicle (UAV/UGV), each of which has certain limitations and constraints. Besides, the robotic systems suffer from inefficient control within cluttered indoor environments containing many obstacles. This paper presents a multi-agent robotic system (MARS) for automatic UAV-UGV path planning and indoor navigation to automate sensory data collection. The proposed MARS consists of a new system architecture that defines the attributes and data requirements for UAV and UGV indoor path planning. To improve indoor navigation in cluttered environments, an enhanced shunting short-term memory model is established to optimize the pathfinding of UAV/UGV for data collection. Assessment of indoor navigation is conducted with a simulation-based approach and LiDAR SLAM. A mediating agent, which harnesses a control algorithm and information exchange mechanism, is proposed to interoperate UAV and UGV for automated data collection. The proposed new MARS is examined in experiments, in which a single UAV, dual UAVs, and combined UAV-UGV are tested in a research laboratory. The result indicates that the MARS can support automated path planning and indoor navigation for 2D image and 3D point cloud data collection.
| Type: | Article |
|---|---|
| Title: | Multi-agent robotic system (MARS) for UAV-UGV path planning and automatic sensory data collection in cluttered environments |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1016/j.buildenv.2022.109349 |
| Publisher version: | https://doi.org/10.1016/j.buildenv.2022.109349 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Building automation, Sensory data, Multi-agent system, Robotics, Indoor inspection, Unmanned aerial/ground vehicle |
| UCL classification: | UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment UCL > Provost and Vice Provost Offices > UCL BEAMS UCL |
| URI: | https://discovery-pp.ucl.ac.uk/id/eprint/10151267 |
Archive Staff Only
 |
View Item |
