5 Conclusions
Based on the deep learning, the waypoint behavior of USV has been taken as the
research object. The speed, steering angle, capture radius, slip radius, lead distance, and
lead damper have been selected as the features to establish the trajectory tracking
control model of the waypoint behavior. By using the experimental sample data
obtained from MOOS-IvP for training, the following conclusions have been drawn:
(1) Based on the deep learning method, we can effectively solve the complex problems brought by the environment, equipment, and algorithms faced by the USV.
(2) The prediction model can judge the influence of different feature parameters on
the waypoint behavior of the USV, and the prediction accuracy can reach 84.9%.
References
1. Švec P, Thakur A (2014) Target following with motion prediction for unmanned surface
vehicle operating in cluttered environments. Auton Robot 36:383–405
2. LaValle SM (2006) Planning algorithms. Cambridge University Press, Cambridge. http://
planning.cs.uiuc.edu
3. Gupta A, Novitzky M, Benjamin M (2018) Learning autonomous marine behaviors in
MOOS-IvP. In: Oceans 2018 MTS/IEEE Charleston
4. Li C (2016) Adaptive RBF neural network control for unmanned surface vessel course
tracking. In: Sixth international conference on information science and technology
5. Wang Y, Tong J, Song TY, Wan ZH (2018) Unmanned surface vehicle course tracking
control based on neural network and deep deterministic policy gradient algorithm. In:
Oceans-MTS/IEEE Kobe Techno-Oceans (OTO)
6. Moe S, Rustad AM, Hanssen KG (2018) Machine learning in control systems: an overview
of the state of the art
7. Xu QY, Yang YQ, Zhang CJ, Zhang L (2018) Deep convolutional neural network-based
autonomous marine vehicle maneuver. Int J Fuzzy Syst 20(2):687–699
8. Chen YZ, Shi YY, Zhang BS (2019) Optimal control via neural networks: a convex
approach. To Appear in International conference on learning representations
9. Marius B (2019) End-to-end deep learning for self-driving cars. http://devblogs.nvidia.com/
deep-learning-self-driving-cars
10. Benjamin MR, Newman P, Schmidt H, Leonard JJ (2009) An overview of MOOS-IvP and a
brief users guide to the IvP helm autonomy software. MIT Department of Mechanical
Engineering, Technical report MIT-CSAIL-TR-2009-028
598
W. Sun and X. Gao
Based on the deep learning, the waypoint behavior of USV has been taken as the
research object. The speed, steering angle, capture radius, slip radius, lead distance, and
lead damper have been selected as the features to establish the trajectory tracking
control model of the waypoint behavior. By using the experimental sample data
obtained from MOOS-IvP for training, the following conclusions have been drawn:
(1) Based on the deep learning method, we can effectively solve the complex problems brought by the environment, equipment, and algorithms faced by the USV.
(2) The prediction model can judge the influence of different feature parameters on
the waypoint behavior of the USV, and the prediction accuracy can reach 84.9%.
References
1. Švec P, Thakur A (2014) Target following with motion prediction for unmanned surface
vehicle operating in cluttered environments. Auton Robot 36:383–405
2. LaValle SM (2006) Planning algorithms. Cambridge University Press, Cambridge. http://
planning.cs.uiuc.edu
3. Gupta A, Novitzky M, Benjamin M (2018) Learning autonomous marine behaviors in
MOOS-IvP. In: Oceans 2018 MTS/IEEE Charleston
4. Li C (2016) Adaptive RBF neural network control for unmanned surface vessel course
tracking. In: Sixth international conference on information science and technology
5. Wang Y, Tong J, Song TY, Wan ZH (2018) Unmanned surface vehicle course tracking
control based on neural network and deep deterministic policy gradient algorithm. In:
Oceans-MTS/IEEE Kobe Techno-Oceans (OTO)
6. Moe S, Rustad AM, Hanssen KG (2018) Machine learning in control systems: an overview
of the state of the art
7. Xu QY, Yang YQ, Zhang CJ, Zhang L (2018) Deep convolutional neural network-based
autonomous marine vehicle maneuver. Int J Fuzzy Syst 20(2):687–699
8. Chen YZ, Shi YY, Zhang BS (2019) Optimal control via neural networks: a convex
approach. To Appear in International conference on learning representations
9. Marius B (2019) End-to-end deep learning for self-driving cars. http://devblogs.nvidia.com/
deep-learning-self-driving-cars
10. Benjamin MR, Newman P, Schmidt H, Leonard JJ (2009) An overview of MOOS-IvP and a
brief users guide to the IvP helm autonomy software. MIT Department of Mechanical
Engineering, Technical report MIT-CSAIL-TR-2009-028
598
W. Sun and X. Gao