Abstract

This paper focuses on a service robot equipped with a manipulator to take the elevator autonomously. When the robot has no communication with the elevator, the robot multi-floor navigation needs to detect and operate the unknown elevator button like a human. This paper researches on the elevator button detection and the elevator button operation. First, a button detection algorithm is proposed based on stroke width transform for regions extraction, and then these regions are recognized by the convolutional neural network. Then, this paper proposes an algorithm to plan the operation pose of the manipulator based on the plane of the target button. Finally, the experiments show that the proposed algorithm can detect and operate the unknown elevator panel button.

Full Text
Published version (Free)

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 call