Sensor‐driven autonomous underwater inspections: A receding‐horizon RRT‐based view planning solution for AUVs

Abstract

Autonomous Underwater Vehicles (AUVs) are used by the scientific community for various applications, from collecting well-distributed high-quality data to mapping the seafloor or exploring unknown areas. Nonpredictable environmental conditions and sensor acquisitions make the design of AUV surveys challenging even for expert operators. Multiple attempts are required, and the collected data quality is not guaranteed: The AUV passively stores the sensors' acquisitions that are then analyzed offline after its recovery. In Forward-Looking SONAR (FLS) seabed inspections, the vehicle follows lawnmower paths designed by an expert operator that considers the sensor characteristics. The performance of FLSs is affected by several environmental conditions and possible protruding objects. This paper presents a probabilistic framework for FLS-based seabed inspections that endow the AUV with the ability to autonomously conducting the survey and ensure adequate coverage of the target area. A three-dimensional probabilistic occupancy mapping system for FLS reconstructions to update the covered area map was developed. The map is used by the Coverage Path Planning (CPP) algorithm to evaluate the visibility of the viewpoints that are generated as nodes of a random tree. The Next-Best Viewpoint (NBV) is selected as the first node in the branch expected to collect more data, and the path to reach the NBV is computed using the rapidly exploring random tree (RRT*) algorithm. The sensor-driven coverage approach is used in a receding-horizon manner. The proposed Receding-Horizon Coverage Approach was validated with simulations and real prerecorded data. Finally, the framework was used online during an experimental campaign where several FLS seabed inspections were performed.

AUV planning and calibration method considering concealment in uncertain environments

“…Another approach proposed by Han et al (2020) is the CCPP scheme for underwater gliders, which considers both the gliding angle and the relative distance to obstacles when navigating. Zacchini et al (2022) developed a 3D probabilistic occupancy map system for sonar reconstruction with backward horizon sensor-driven coverage adaptable to environmental uncertainty.…”

Recent advances in Rapidly-exploring random tree: A review

“…Some literature design branching mode depends on scenarios; for example, Zacchini et al [ 78 ] predict the effect of taking more actions in unknown environments by considering branch information gain and assessing the impact of the rewiring process on tree growth. To enable the spacecraft to plan time-optimal attitude trajectories under complex pointing constraints, Xu et al [ 79 ] propose an intelligent branching model based on geometric-level spherical and quadrilateral interpolation, which is capable of planning suboptimal solutions within a few milliseconds.…”

“…In subsea inspection tasks, traditional strategies utilize sensors to passively collect data; these paths can be optimized by taking into account the characteristics of the sensors, but cannot predict the environmental conditions, therefore, they do not guarantee access to data across the entire domain of interest and may lead to repeated attempts. To address this problem, Leonardo et al [ 78 ] build the forward‐looking sonar seabed inspections framework, present a RRT-based sensor-driven receding horizon approach, and introduce the entropy-based evaluation metric for solution searching. Comparative results show the proposed algorithm achieves the same coverage area rate at a shorter path cost.…”

“…The need to model unknown scenarios is largely due to GPS denial, in which case many researchers also load sensors on robots to build maps, such as sonar for seabed inspection [ 78 , 238 ] and tunnel inspection [ 233 ], LiDAR for indoor inspection [ 201 ], tunnel inspection [ 233 ] and fire guidance [ 202 ], and depth camera for indoor scene modeling [ 204 ]. This environment exploration strategy has been proven to be effective, for example, three UAVs each loaded with a depth camera take 209.4s to explore a 10 m*8 m*3 m room filled with various types of obstacles [ 204 ], whereas underwater robots loaded with sonar sensors detect more than 90 % of the 67 m*67 m*4 m seafloor area in a few minutes [ 238 ], but there are problems such as difficulty in selecting interest point regions, unstable modeling quality, and difficult trade-off between modeling accuracy and efficiency.…”

Scanning Scheme for Underwater High-Rise Pile Cap Foundation Based on Imaging Sonar

“…Shipping vessels are the most common carriers of IS [13], but it is difficult for the imaging accuracy to meet the demand because ships are often scanned tens of meters away to prevent collision with piles. AUVs can be operated remotely and approach the target and have been successfully applied in breakwater structure scanning [23], subsea pipeline inspection [24], and subsea topography scanning [25]. Song [26] applied AUVs to inspect the Jiaozhou Bay Bridge in Qingdao, China.…”