Abstract
In this paper, a Next Best View (NBV) approach with a profiling stage and a novel utility function for 3D reconstruction using an Unmanned Aerial Vehicle (UAV) is proposed. The proposed approach performs an initial scan in order to build a rough model of the structure that is later used to improve coverage completeness and reduce flight time. Then, a more thorough NBV process is initiated, utilizing the rough model in order to create a dense 3D reconstruction of the structure of interest. The proposed approach exploits the reflectional symmetry feature if it exists in the initial scan of the structure. The proposed NBV approach is implemented with a novel utility function, which consists of four main components: information theory, model density, traveled distance, and predictive measures based on symmetries in the structure. This system outperforms classic information gain approaches with a higher density, entropy reduction and coverage completeness. Simulated and real experiments were conducted and the results show the effectiveness and applicability of the proposed approach.
Highlights
The ability to reconstruct 3D models of large structures has been utilized in various applications such as infrastructure inspection [1,2,3] and search and rescue [4,5]
To navigate iteratively in an informed manner, the Best View (NBV) is determined based on the data gathered, and a utility function that evaluates the gain from different candidate locations [10,11,12]
Two real indoor experiments were conducted using different structures in order to validate the applicability of the Guided Next Best View (NBV) system
Summary
The ability to reconstruct 3D models of large structures has been utilized in various applications such as infrastructure inspection [1,2,3] and search and rescue [4,5]. Reconstruction tasks are typically performed using conventional methods either by following an autonomous procedure such as moving a sensor around the structure of interest at fixed waypoints or following manual procedure. Performing this process manually is considered time consuming and provides no guarantees of coverage completeness or reproducibility. If an existing model is provided, an exploration path could be computed offline, which is executed by the robot during the reconstruction task [6,7,8,9] These reference models are usually not available, not provided, or inaccurate. To navigate iteratively in an informed manner, the Best View (NBV) is determined based on the data gathered, and a utility function that evaluates the gain from different candidate locations (usually referred to by viewpoints) [10,11,12]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have