Design of an obstacle detection system for marine autonomous vehicles
A. Sorbara, Angelo Odetti, Marco Bibuli, Enrica Zereik, Gabriele Bruzzone
- 发表年份
- 2015
- 引用次数
- 7
摘要
The need for the transfer of robotic technology from the research laboratories to civilian applications is now a major issue for the robotic community. In the field of marine robotics is particularly important to endow the robot with a safe and highly reliable Navigation Guidance and Control (NGC) system which enables the Unmanned Surface Vehicle (USV) to safely navigate even in presence of human activities such as commercial and recreational traffic, swimmers, rowers, etc. A key aspect to achieve a safe autonomous navigation is the ability of the robot to recognize well in advance the presence of an unexpected, potentially moving, obstacle. This function represents the base brick for the development of a collision avoidance systems smart enough to reactively detect unexpected obstacles and perform the necessary avoidance maneuvers to safely prevent collisions. The present paper describes the design of an a innovative obstacle detection sensor, combining both passive and active optical devices and based on a new concept of optronic system. It is specifically conceived for collision avoidance tasks in marine environments, designed to be easily mounted on small-medium sized USVs. Its innovation consists in the interaction between the different integrated sensors, that are in fact totally decoupled. The paper presents the functional architecture of the object detection sensor together with the preliminary mechanical design. Moreover some experimental data collected by the sensor are reported, and some simulations, highlighting the ability of the system to detect and correctly avoid both still obstacles and mobile traversing obstacles, are shown.
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002