Learning an AUV docking maneuver with a convolutional neural network

Sans-Muntadas, Albert; Kelasidi, Eleni; Pettersen, Kristin Y.; Brekke, Edmund

doi:10.1016/j.ifacsc.2019.100049

Cited by 12 publications

(7 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This system is called end-to-end because instead of using lane detection, path planning and control algorithms separately, it optimizes all this processing simultaneously using the representation model learned by CNN. Refs [13,15,16,[18][19][20][21][22][23] could be classified as end-to-end methods.…”

Section: State Of the Art In Visual Servoingmentioning

confidence: 99%

A Deep Neural Network Sensor for Visual Servoing in 3D Spaces

Durdevic

Ortiz-Arroyo

2020

Sensors

View full text Add to dashboard Cite

This paper describes a novel stereo vision sensor based on deep neural networks, that can be used to produce a feedback signal for visual servoing in unmanned aerial vehicles such as drones. Two deep convolutional neural networks attached to the stereo camera in the drone are trained to detect wind turbines in images and stereo triangulation is used to calculate the distance from a wind turbine to the drone. Our experimental results show that the sensor produces data accurate enough to be used for servoing, even in the presence of noise generated when the drone is not being completely stable. Our results also show that appropriate filtering of the signals is needed and that to produce correct results, it is very important to keep the wind turbine within the field of vision of both cameras, so that both deep neural networks could detect it.

show abstract

Section: State Of the Art In Visual Servoingmentioning

confidence: 99%

A Deep Neural Network Sensor for Visual Servoing in 3D Spaces

Durdevic

Ortiz-Arroyo

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…For the AUV docking, a number of studies about the detection position of AUV in short distance have been performed so far [8][9][10][11]. Normally, they use cameras with pinhole camera method to determine the position of AUV and station as in [12] and [13].…”

Section: Introductionmentioning

confidence: 99%

On the Position Determination of Docking Station for AUVs Using Optical Sensor and Neural Network

Nhat

Choi

et al. 2020

Int. j. eng. technol. innov.

View full text Add to dashboard Cite

Detecting the relative position of the docking station is a very important issue for the homing of AUVs (Autonomous Unmanned Vehicles). To detect the position of the light source, a pinhole camera model structure was proposed like the camera model. However, due to the sensor resolution and the distortion errors of the pinhole camera system, the application of the camera of docking the under turbid sea environments is almost impossible. In this paper, a new method detecting the position of the docking station using a light source is presented. Also, a newly developed optical sensor which makes it much easier to sense the light source than the camera system for homing of the AUV under the water is performed. In addition, to improve the system, a neural network (NN) algorithm constructing a model relating the light inputs and optical sensor which are developed in this study is proposed. To evaluate the performance of the NN algorithm, the experiments were performed in the air beforehand. The result shows that the NN algorithm with AUV docking system using the NN model is better than the pinhole camera model.

show abstract

“…In most of these studies, advanced simultaneous localisation and mapping (SLAM) computer-vision strategies are coupled with line-of-sight guidance and classical control algorithms, such as proportional-integral-derivative (PID) and sliding-mode (SM) control. Alternatively, the application of deep learning for the the overall control of the AUV for the docking manoeuvre was investigated by Sans-Muntadas et al [23].…”

Section: Introductionmentioning

confidence: 99%

Docking Control of an Autonomous Underwater Vehicle Using Reinforcement Learning

2019

View full text Add to dashboard Cite

To achieve persistent systems in the future, autonomous underwater vehicles (AUVs) willneed to autonomously dock onto a charging station. Here, reinforcement learning strategies wereapplied for the first time to control the docking of an AUV onto a fixed platform in a simulationenvironment. Two reinforcement learning schemes were investigated: one with continuous stateand action spaces, deep deterministic policy gradient (DDPG), and one with continuous state butdiscrete action spaces, deep Q network (DQN). For DQN, the discrete actions were selected as stepchanges in the control input signals. The performance of the reinforcement learning strategies wascompared with classical and optimal control techniques. The control actions selected by DDPG sufferfrom chattering effects due to a hyperbolic tangent layer in the actor. Conversely, DQN presents thebest compromise between short docking time and low control effort, whilst meeting the dockingrequirements. Whereas the reinforcement learning algorithms present a very high computational costat training time, they are five orders of magnitude faster than optimal control at deployment time,thus enabling an on-line implementation. Therefore, reinforcement learning achieves a performancesimilar to optimal control at a much lower computational cost at deployment, whilst also presentinga more general framework.

show abstract

Learning an AUV docking maneuver with a convolutional neural network

Cited by 12 publications

References 12 publications

A Deep Neural Network Sensor for Visual Servoing in 3D Spaces

A Deep Neural Network Sensor for Visual Servoing in 3D Spaces

On the Position Determination of Docking Station for AUVs Using Optical Sensor and Neural Network

Docking Control of an Autonomous Underwater Vehicle Using Reinforcement Learning

Contact Info

Product

Resources

About