1. Building Metric-Topological Map to Efficient Object Search for Mobile Robot
- Author
-
Ying Zhang, Guohui Tian, Xuyang Shao, Shaopeng Liu, Mengyang Zhang, and Peng Duan
- Subjects
business.industry ,Computer science ,Mobile robot ,Construct (python library) ,Object (computer science) ,Viewing angle ,Control and Systems Engineering ,Position (vector) ,Metric (mathematics) ,Robot ,Computer vision ,Topological map ,Artificial intelligence ,Electrical and Electronic Engineering ,business - Abstract
In this paper, an effective solution is presented for efficient robotic object search by leveraging metric-topological map. Based on the reference of objects, we construct a novel metric-topological map that is particularly suitable for object search, where the topological nodes represent the object-related locations in the environment. To effectively determine the creation of nodes on the map, a rapid adjustment method of the robot viewing angle is introduced. Compared with traditional maps, the map built in this paper not only removes redundant nodes irrelevant to objects, but also facilitates the mobile robot to locate the target object. Besides, inspired by human search behavior, a search strategy is proposed by considering the tradeoff between the current position of the robot and the distance from the robot to the promising location of finding the target object, in order to further improve the search efficiency. The effectiveness and efficiency of our solution is verified through comparative experiments in both simulation and real-world environments. Results show that the presented approach always allows the mobile robot to find the target object efficiently and reliably, while achieving human-like search behavior.
- Published
- 2022
- Full Text
- View/download PDF