A Shadowcasting-Based Next-Best-View Planner for Autonomous 3D Exploration

Batinović, Ana; Ivanović, Antun; Petrović, Tamara; Bogdan, Stjepan

doi:10.1109/lra.2022.3146586

Cited by 21 publications

(14 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other side, other works select the NBV after evaluating an objective function on a list of randomly chosen candidates in the neighbourhood of the present point. References [44][45][46] use a fitness function which takes into account the potential information gain (measuring the free space around the candidate) and the cost of getting to it from the current location. In the objective function of [47], free space, and information gain (measured image entropy) are taken into account and a penalty is added to points already visited.…”

Section: Nbv Techniquesmentioning

confidence: 99%

“…Equation ( 1) is similar in its external form to [47]. However, while the latter considers free space and potential information gain and penalises visited points, the model presented here considers, for example, [44] and [46], the distance to the candidate. A pseudocode implementation of the function can be found in algorithm 5.…”

Section: System Integration 341 Nbv Selection Based On Image-laser Fu...mentioning

confidence: 99%

See 1 more Smart Citation

Active robotic search for victims using ensemble deep learning techniques

García-Samartín,

Cruz Ulloa,

del Cerro

et al. 2024

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

In recent years, legged quadruped robots have proved to be a valuable support to humans in dealing with Search and Rescue (SAR) operations. These robots can move with great ability in complex terrains, unstructured environments or regions with many obstacles. This work employs the quadruped robot ARTU-R (A1 Rescue Tasks UPM Robot) by Unitree, equipped with an RGB-D camera and a lidar, to perform victim searches in post-disaster scenarios. Exploration is done not by following a pre-planned path (as common methods) but by prioritising the areas most likely to harbour victims. To accomplish that task, both Indirect Search (IS) and Next Best View (NBV) techniques have been used. When ARTU-R gets inside an unstructured and unknown environment, it selects the next exploration point from a series of candidates. This operation is performed by comparing, for each candidate, the distance to reach it, the unexplored space around it and the probability of a victim being in its vicinity. This probability value is obtained using a Random Forest, which processes the information provided by a Convolutional Neural Network (CNN). Unlike other AI techniques, random forests are not black box models; humans can understand their decision-making processes. The system, once integrated, achieves speeds comparable to other state-of-the-art algorithms in terms of exploration, but concerning victim detection, the tests show that the resulting smart exploration generates logical paths --from a human point of view-- and that ARTU-R tends to move first to the regions where victims are present.

show abstract

Section: Nbv Techniquesmentioning

confidence: 99%

Section: System Integration 341 Nbv Selection Based On Image-laser Fu...mentioning

confidence: 99%

Active robotic search for victims using ensemble deep learning techniques

García-Samartín,

Cruz Ulloa,

del Cerro

et al. 2024

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…We consider the environment to be known, either given through a known map or obtained by an exploration or mapping algorithm, i.e. as proposed in [33]. In order to find an obstacle-free path in the environment, we employ the wellknown RRT* algorithm [34], although other path planners can be used as well.…”

Section: A Parabolic Free-fall Trajectorymentioning

confidence: 99%

Parabolic Airdrop Trajectory Planning for Multirotor Unmanned Aerial Vehicles

Ivanović

Orsag

2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

In this paper we present a motion planning method for the parabolic airdrop using multirotor UAVs. The first step is to determine a set of parabolic trajectories that can reach a desired target. Next, we plan a path to the payload launch point, and based on that path we plan a collision-free trajectory that safely executes both launch and stopping motions. The described motion planning method is extensively tested in the simulation environment: by executing many trajectories through repeatability analysis; planning in a large city-like environment; and planning in a dense office environment. Furthermore, the method is tested through real-world experiments in indoor and outdoor environments. In both environments, we performed the repeatability analysis and the obstacle avoidance experiments while delivering the payload to the target. The video demonstrating our simulation and experimental analyses is available at: https://youtu.be/HJCH2vJEuuo INDEX TERMS Unmanned aerial vehicles, motion planning, unmanned autonomous systems.

show abstract

“…Thereby, a feasible path between the viewpoints is computed, avoiding obstacles and adhering to robot constraints. 8,9 As depicted in Figure 2, the whole coverage path planning procedure can be split up into the five steps: virtual model, preparation of the input model; viewpoint generation, determination of viewpoint candidates; viewpoint selection, removal of unnecessary viewpoint candidates; path planning, ordering of the remaining viewpoints; inspection, the actual measurement process.…”

Section: View Planningmentioning

confidence: 99%

“…Thereby, a feasible path between the viewpoints is computed, avoiding obstacles and adhering to robot constraints. 8,9…”

Section: Introductionmentioning

confidence: 99%

Model based view planning for the robot-guided automation of optical 3D digitization based on a variable mesh resolution processing approach

Geiger

Sauerbier

Montavon

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part B:

View full text Add to dashboard Cite

The digital 3D representation of manufactured parts plays a crucial role in the quality assurance of small, individualized lot sizes. Machine vision systems, such as optical 3D scanners, guided by an industrial robotic arm, allow for a contactless full digital reconstruction of surface geometries. In order to digitize the whole geometry, it is necessary to acquire 3D scans from multiple viewpoints with respect to the part, which in combination cover the entire surface. With efficiency in mind, this results in an optimization problem between a high surface area coverage and low measurement effort, referred to as the view planning problem. In the presented work, two popular viewpoint candidate generation methods are implemented: Firstly, a surface-based random sampling method, which generates viewpoints within a solution space, in which visibility of a given model surface can be expected. Secondly, a view sphere viewpoint generation approach, which is independent of the object geometry but avoids clustering by generating evenly spaced viewpoints on a sphere around the centre of the object. Using an adjustable remeshing procedure, a multi-stage approach is implemented by generating multiple meshes with different resolutions. Through this, the benefits of working on a coarse mesh, such as fast viewpoint candidate evaluation and selection, are combined with the level of detail of a fine mesh. It is found that it is possible to considerably reduce the mesh resolution while maintaining a reasonably high surface area coverage on a reference model. Applying the proposed procedure to the view planning for a state-of-the-art 3D fringe light projection scanner with highly sophisticated scanning capabilities, it is demonstrated that the view sphere approach is more suitable for this use-case due to the large measurement volumes of the 3D scanner. The frequently used random sampling approach requires an excessively higher computational effort to achieve similar results in comparison.

show abstract

A Shadowcasting-Based Next-Best-View Planner for Autonomous 3D Exploration

Cited by 21 publications

References 29 publications

Active robotic search for victims using ensemble deep learning techniques

Active robotic search for victims using ensemble deep learning techniques

Parabolic Airdrop Trajectory Planning for Multirotor Unmanned Aerial Vehicles

Model based view planning for the robot-guided automation of optical 3D digitization based on a variable mesh resolution processing approach

Contact Info

Product

Resources

About