Abstract
Micro aerial vehicles (MAVs) are an excellent platform for autonomous exploration. Most MAVs rely mainly on cameras for buliding a map of the 3D environment. Therefore, vision-based MAVs require an efficient exploration algorithm to select viewpoints that provide informative measurements. In this paper, we propose an exploration approach that selects in real time the next-best-view that maximizes the expected information gain of new measurements. In addition, we take into account the cost of reaching a new viewpoint in terms of distance and predictability of the flight path for a human observer. Finally, our approach selects a path that reduces the risk of crashes when the expected battery life comes to an end, while still maximizing the information gain in the process. We implemented and thoroughly tested our approach and the experiments show that it offers an improved performance compared to other state-of-the-art algorithms in terms of precision of the reconstruction, execution time, and smoothness of the path.
Highlights
When performing autonomous 3D reconstruction, the exploration strategy determines the efficiency with which an accurate 3D model of the environment can be obtained
We implemented our approach in C++ using Robot Operating System (ROS) and tested it in simulation as well as in a real indoor environment, along with other state-of-the-art algorithms
Our experiments are designed to show the capabilities of our method and to support the key claims we made in the introduction, which are: (i) our approach yields a map with a low uncertainty in the probabilistic model; (ii) it avoids abrupt changes of direction during the flight; (iii) it does not generate a longer path by taking into account the aforementioned aspects; and (iv) it is able to compute the best viewpoint in real time and online on an exploring system
Summary
When performing autonomous 3D reconstruction, the exploration strategy determines the efficiency with which an accurate 3D model of the environment can be obtained. The problem of exploration consists of selecting the best viewpoints to cover the environment with the available sensors to obtain an accurate 3D model. When no a priori information about the environment is available, a popular approach to this problem is the iterative selection of the next-best-view. This approach consists in selecting online the pose for the sensor that best satisfies certain criteria, usually related to the amount of information acquired by the new observations and the cost of executing the action. Since no information about the environment is initially available, this kind of approach works in a greedy fashion, i.e., it is executed online during the mission and considers, at each iteration, the new measurements acquired by the sensor to plan the pose. Each voxel contains all the data needed for computing the next-best-view and has three possible states: unknown, free or occupied
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
