Optimal Online Searching with an Autonomous Robot

Title: Optimal Online Searching with an Autonomous Robot

Abstract: This research investigates the problem of online searching for an object hidden behind a corner using a real autonomous robot named Kurt3D. The robot uses a three-dimensional laser scanner to build a virtual three-dimensional environment, helping it plan trajectories, avoid obstacles, and identify objects. The study derives a practically useful and asymptotically optimal strategy that guarantees a competitive ratio of 2, which is a significant improvement over previous studies. This strategy is applied by Kurt3D and documented in a separate video.

Main Research Question: How can we develop an optimal online searching strategy for an autonomous robot to locate an object hidden behind a corner, considering stop-and-scan limitations?

Methodology: The study uses a real autonomous robot, Kurt3D, which employs a three-dimensional laser scanner to survey its environment. The researchers derive an optimal online searching strategy that minimizes the total time to locate an object or explore a region, taking into account both travel time and scan time.

Results: The study presents a practically useful and asymptotically optimal strategy that guarantees a competitive ratio of 2. This strategy is applied by Kurt3D and demonstrated in a separate video.

Implications: This research highlights the need for closer interaction between theoretical studies and practical applications. It also opens up new opportunities for algorithmic research in autonomous robotics and online searching.

Link to Article: https://arxiv.org/abs/0404036v1 Authors: arXiv ID: 0404036v1