Title Highly Maneuverable Ground Reconnaissance Robot Based on Machine Learning
Authors Wang, Peixuan
Peng, Xiaorui
Chen, Jiayu
Chen, Jiacheng
Affiliation Beijing Inst Technol, Sch Automat, Beijing, Peoples R China.
Beijing Inst Technol, Sch Informat & Elect, Beijing, Peoples R China.
Peking Univ, Coll Engn, Beijing, Peoples R China.
Keywords recon robot
machine learning
double wishbone independent suspension
SVM
Mecanum wheel
Issue Date 2018
Publisher 2018 3RD INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION ENGINEERING (ICRAE)
Citation 2018 3RD INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION ENGINEERING (ICRAE). 2018, 102-106.
Abstract This article mainly introduces the application of a reconnaissance robot's mechanical design, motion control algorithm and image processing program. In the mechanical structure, it adopts a motor system that drives a Mecanum wheel with double wishbone independent suspension to ensure that the robot has high speed and high mobility. The robot is equipped with a two-degree-of-freedom pan/tilt for installing a wireless video camera, which can acquire images with free viewing angles to facilitate remote control of operators and obtain on-site intelligence. The image processing program uses the machine learning method of the HOG feature and the SVM classifier to identify target objects such as enemy forces or certain types of explosive objects in disaster scenarios to reduce the pressure on operators and increase the reaction speed.
URI http://hdl.handle.net/20.500.11897/571589
Indexed CPCI-S(ISTP)
Appears in Collections: 工学院

Files in This Work
There are no files associated with this item.

Web of Science®


0

Checked on Last Week

百度学术™


0

Checked on Current Time




License: See PKU IR operational policies.