Predicting the Trajectory Tracking
Control of Unmanned Surface Vehicle
Based on Deep Learning
Wenli Sun
1(&) and Xu Gao
2
1 Navigation College, Dalian Maritime University, Dalian 116026, China
itslab@dlmu.edu.cn
2 Dalian Maritime Navigation Co., LTD., Dalian 116026, China
Abstract. The waypoint behavior is a method of trajectory tracking control for
unmanned surface vehicle (USV). We take the waypoint behavior as the
research object, choose speed, steering angle, capture radius, slip radius, lead
distance, lead damper as features, and establish a prediction model for trajectory
tracking control of USV based on deep neural network. The model effectively
predicts the navigation effect and provides assistance and reference for the
maneuvering decision of the USV. Test results show that the predicted results
and the reference samples have same tendency. The proposed model can
improve the performance of trajectory tracking control for the USV.
Keywords: USV Á Deep learning Á Waypoint behavior Á Trajectory tracking
control
1 Introduction
In recent years, unmanned surface vehicle (USV) has been widely used in military and
civilian fields, playing an important role in port defense, maritime safety, marine
surveying, and mapping. For USV, autonomous control technology is a core issue. It is
the key performance that USV can execute tasks such as trajectory tracking, automatic
collision avoidance, and cooperative formation [1]. Among them, the trajectory
tracking control is a broad and important research area [2].
Many researches have used neural network in USV’s control problem, particularly for
trajectory tracking control, instead of traditional control systems which are based on
identifying and modeling a nonlinear and complicated system. Arjun Gupta et al. adopt
deep reinforcement learning to learn autonomous behaviors without relying on explicit
models [3]. Cunhe Li et al. present an adaptive radial basis function neural network
controller for nonlinear course control of USV, which can enable the USV to track the
designed heading angle and turning rate [4]. A neural network is used to control the USV
for course tracking by Yan Wang et al., solving the uncertainty problem of ship motion
[5]. Signe Moe et al. train a neural network to mimic a feedback linearizing speed
controller for an autonomous ship, which outperforms the traditional controller in terms of
modeling errors and measurement noise [6]. A deep neural network is used to learn the
automatic maneuver features of the USV based on vision system by Xu et al. [7].
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 591–598, 2020.
https://doi.org/10.1007/978-981-15-0187-6_70
Control of Unmanned Surface Vehicle
Based on Deep Learning
Wenli Sun
1(&) and Xu Gao
2
1 Navigation College, Dalian Maritime University, Dalian 116026, China
itslab@dlmu.edu.cn
2 Dalian Maritime Navigation Co., LTD., Dalian 116026, China
Abstract. The waypoint behavior is a method of trajectory tracking control for
unmanned surface vehicle (USV). We take the waypoint behavior as the
research object, choose speed, steering angle, capture radius, slip radius, lead
distance, lead damper as features, and establish a prediction model for trajectory
tracking control of USV based on deep neural network. The model effectively
predicts the navigation effect and provides assistance and reference for the
maneuvering decision of the USV. Test results show that the predicted results
and the reference samples have same tendency. The proposed model can
improve the performance of trajectory tracking control for the USV.
Keywords: USV Á Deep learning Á Waypoint behavior Á Trajectory tracking
control
1 Introduction
In recent years, unmanned surface vehicle (USV) has been widely used in military and
civilian fields, playing an important role in port defense, maritime safety, marine
surveying, and mapping. For USV, autonomous control technology is a core issue. It is
the key performance that USV can execute tasks such as trajectory tracking, automatic
collision avoidance, and cooperative formation [1]. Among them, the trajectory
tracking control is a broad and important research area [2].
Many researches have used neural network in USV’s control problem, particularly for
trajectory tracking control, instead of traditional control systems which are based on
identifying and modeling a nonlinear and complicated system. Arjun Gupta et al. adopt
deep reinforcement learning to learn autonomous behaviors without relying on explicit
models [3]. Cunhe Li et al. present an adaptive radial basis function neural network
controller for nonlinear course control of USV, which can enable the USV to track the
designed heading angle and turning rate [4]. A neural network is used to control the USV
for course tracking by Yan Wang et al., solving the uncertainty problem of ship motion
[5]. Signe Moe et al. train a neural network to mimic a feedback linearizing speed
controller for an autonomous ship, which outperforms the traditional controller in terms of
modeling errors and measurement noise [6]. A deep neural network is used to learn the
automatic maneuver features of the USV based on vision system by Xu et al. [7].
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 591–598, 2020.
https://doi.org/10.1007/978-981-15-0187-6_70