Autonomous underwater vehicle control using reinforcement learning policy search methods

Autonomous underwater vehicles (AUV) represent a challenging control problem with complex, noisy, dynamics. Nowadays, not only the continuous scientific advances in underwater robotics but the increasing number of subsea missions and its complexity ask for an automatization of submarine processes. This paper proposes a high-level control system for solving the action selection problem of an autonomous robot. The system is characterized by the use of reinforcement learning direct policy search methods (RLDPS) for learning the internal state/action mapping of some behaviors. We demonstrate its feasibility with simulated experiments using the model of our underwater robot URIS in a target following task

A mission control system for an autonomous underwater vehicle

The presented work focuses on the theoretical and practical aspects concerning the design and development of a formal method to build a mission control system for autonomous underwater vehicles bringing systematic design principles for the formal description of missions using Petri nets. The proposed methodology compounds Petri net building blocks within it to de_ne a mission plan for which it is proved that formal properties, such as reachability and reusability, hold as long as these same properties are also guaranteed by each Petri net building block. To simplify the de_nition of these Petri net blocks as well as their composition, a high level language called Mission Control Language has been developed. Moreover, a methodology to ensure coordination constraints for teams of multiple robots as well as the de_nition of an interface between the proposed system and an on-board planner able to plan/replan sequences of prede_ned mission plans is included as well. Results of experiments with several real underwater vehicles and simulations involving an autonomous surface craft and an autonomous underwater vehicles are presented to show the system's capabilities.

Using petri nets to specify and execute missions for autonomous underwater vehicles

This paper presents the design and implementation of a mission control system (MCS) for an autonomous underwater vehicle (AUV) based on Petri nets. In the proposed approach the Petri nets are used to specify as well as to execute the desired autonomous vehicle mission. The mission is easily described using an imperative programming language called mission control language (MCL) that formally describes the mission execution thread. A mission control language compiler (MCL-C) able to automatically translate the MCL into a Petri net is described and a real-time Petri net player that allows to execute the resulting Petri net onboard an AUV are also presented

A behavior-based scheme using reinforcement learning for autonomous underwater vehicles

This paper presents a hybrid behavior-based scheme using reinforcement learning for high-level control of autonomous underwater vehicles (AUVs). Two main features of the presented approach are hybrid behavior coordination and semi on-line neural-Q_learning (SONQL). Hybrid behavior coordination takes advantages of robustness and modularity in the competitive approach as well as efficient trajectories in the cooperative approach. SONQL, a new continuous approach of the Q_learning algorithm with a multilayer neural network is used to learn behavior state/action mapping online. Experimental results show the feasibility of the presented approach for AUVs

An approach to vision-based station keeping for an unmanned underwater vehicle

This paper presents an automatic vision-based system for UUV station keeping. The vehicle is equipped with a down-looking camera, which provides images of the sea-floor. The station keeping system is based on a feature-based motion detection algorithm, which exploits standard correlation and explicit textural analysis to solve the correspondence problem. A visual map of the area surveyed by the vehicle is constructed to increase the flexibility of the system, allowing the vehicle to position itself when it has lost the reference image. The testing platform is the URIS underwater vehicle. Experimental results demonstrating the behavior of the system on a real environment are presented

Policy gradient based Reinforcement Learning for real autonomous underwater cable tracking

This paper proposes a field application of a high-level reinforcement learning (RL) control system for solving the action selection problem of an autonomous robot in cable tracking task. The learning system is characterized by using a direct policy search method for learning the internal state/action mapping. Policy only algorithms may suffer from long convergence times when dealing with real robotics. In order to speed up the process, the learning phase has been carried out in a simulated environment and, in a second step, the policy has been transferred and tested successfully on a real robot. Future steps plan to continue the learning process on-line while on the real robot while performing the mentioned task. We demonstrate its feasibility with real experiments on the underwater robot ICTINEU AUV

Positioning an underwater vehicle through image mosaicking

Mosaics have been commonly used as visual maps for undersea exploration and navigation. The position and orientation of an underwater vehicle can be calculated by integrating the apparent motion of the images which form the mosaic. A feature-based mosaicking method is proposed in this paper. The creation of the mosaic is accomplished in four stages: feature selection and matching, detection of points describing the dominant motion, homography computation and mosaic construction. In this work we demonstrate that the use of color and textures as discriminative properties of the image can improve, to a large extent, the accuracy of the constructed mosaic. The system is able to provide 3D metric information concerning the vehicle motion using the knowledge of the intrinsic parameters of the camera while integrating the measurements of an ultrasonic sensor. The experimental results of real images have been tested on the GARBI underwater vehicle

A review of sensor fusion techniques for underwater vehicle navigation

This work provides a general description of the multi sensor data fusion concept, along with a new classification of currently used sensor fusion techniques for unmanned underwater vehicles (UUV). Unlike previous proposals that focus the classification on the sensors involved in the fusion, we propose a synthetic approach that is focused on the techniques involved in the fusion and their applications in UUV navigation. We believe that our approach is better oriented towards the development of sensor fusion systems, since a sensor fusion architecture should be first of all focused on its goals and then on the fused sensors

Visual SLAM for 3D large-scale seabed acquisition employing underwater vehicles

This paper presents a novel technique to align partial 3D reconstructions of the seabed acquired by a stereo camera mounted on an autonomous underwater vehicle. Vehicle localization and seabed mapping is performed simultaneously by means of an Extended Kalman Filter. Passive landmarks are detected on the images and characterized considering 2D and 3D features. Landmarks are re-observed while the robot is navigating and data association becomes easier but robust. Once the survey is completed, vehicle trajectory is smoothed by a Rauch-Tung-Striebel filter obtaining an even better alignment of the 3D views and yet a large-scale acquisition of the seabed

Gradient-based reinforcement learning techniques for underwater robotics behavior learning

Darrerament, l'interès pel desenvolupament d'aplicacions amb robots submarins autònoms (AUV) ha crescut de forma considerable. Els AUVs són atractius gràcies al seu tamany i el fet que no necessiten un operador humà per pilotar-los. Tot i això, és impossible comparar, en termes d'eficiència i flexibilitat, l'habilitat d'un pilot humà amb les escasses capacitats operatives que ofereixen els AUVs actuals. L'utilització de AUVs per cobrir grans àrees implica resoldre problemes complexos, especialment si es desitja que el nostre robot reaccioni en temps real a canvis sobtats en les condicions de treball. Per aquestes raons, el desenvolupament de sistemes de control autònom amb l'objectiu de millorar aquestes capacitats ha esdevingut una prioritat. Aquesta tesi tracta sobre el problema de la presa de decisions utilizant AUVs. El treball presentat es centra en l'estudi, disseny i aplicació de comportaments per a AUVs utilitzant tècniques d'aprenentatge per reforç (RL). La contribució principal d'aquesta tesi consisteix en l'aplicació de diverses tècniques de RL per tal de millorar l'autonomia dels robots submarins, amb l'objectiu final de demostrar la viabilitat d'aquests algoritmes per aprendre tasques submarines autònomes en temps real. En RL, el robot intenta maximitzar un reforç escalar obtingut com a conseqüència de la seva interacció amb l'entorn. L'objectiu és trobar una política òptima que relaciona tots els estats possibles amb les accions a executar per a cada estat que maximitzen la suma de reforços totals. Així, aquesta tesi investiga principalment dues tipologies d'algoritmes basats en RL: mètodes basats en funcions de valor (VF) i mètodes basats en el gradient (PG). Els resultats experimentals finals mostren el robot submarí Ictineu en una tasca autònoma real de seguiment de cables submarins. Per portar-la a terme, s'ha dissenyat un algoritme anomenat mètode d'Actor i Crític (AC), fruit de la fusió de mètodes VF amb tècniques de PG.

Selective submap joining SLAM for autonomous vehicles

Simultaneous Localization and Mapping (SLAM) do not result in consistent maps of large areas because of gradual increase of the uncertainty for long term missions. In addition, as the size of the map grows the computational cost increases, making SLAM solutions unsuitable for on-line applications. This thesis surveys SLAM approaches paying special attention to those approaches aimed to work on large scenarios. Special focus is given to existing underwater SLAM applications. A technique based on using independent local maps together with a global stochastic map is presented. This technique is called Selective Submap Joining SLAM (SSJS). A global map contains relative transformations between local maps, which are updated once a new loop is detected. Maps sharing several features are fused, maintaining the correlation between landmarks and vehicle's pose. The use of local maps reduces computational costs and improves map consistency as compared to state of the art techniques.

A proposal of a behavior-based control architecture with reinforcement learning for an autonomous underwater robot

Aquesta tesi proposa l'ús d'un seguit de tècniques pel control a alt nivell d'un robot autònom i també per l'aprenentatge automàtic de comportaments. L'objectiu principal de la tesis fou el de dotar d'intel·ligència als robots autònoms que han d'acomplir unes missions determinades en entorns desconeguts i no estructurats. Una de les premisses tingudes en compte en tots els passos d'aquesta tesis va ser la selecció d'aquelles tècniques que poguessin ésser aplicades en temps real, i demostrar-ne el seu funcionament amb experiments reals. El camp d'aplicació de tots els experiments es la robòtica submarina. En una primera part, la tesis es centra en el disseny d'una arquitectura de control que ha de permetre l'assoliment d'una missió prèviament definida. En particular, la tesis proposa l'ús de les arquitectures de control basades en comportaments per a l'assoliment de cada una de les tasques que composen la totalitat de la missió. Una arquitectura d'aquest tipus està formada per un conjunt independent de comportaments, els quals representen diferents intencions del robot (ex.: "anar a una posició", "evitar obstacles",...). Es presenta una recerca bibliogràfica sobre aquest camp i alhora es mostren els resultats d'aplicar quatre de les arquitectures basades en comportaments més representatives a una tasca concreta. De l'anàlisi dels resultats se'n deriva que un dels factors que més influeixen en el rendiment d'aquestes arquitectures, és la metodologia emprada per coordinar les respostes dels comportaments. Per una banda, la coordinació competitiva és aquella en que només un dels comportaments controla el robot. Per altra banda, en la coordinació cooperativa el control del robot és realitza a partir d'una fusió de totes les respostes dels comportaments actius. La tesis, proposa un esquema híbrid d'arquitectura capaç de beneficiar-se dels principals avantatges d'ambdues metodologies. En una segona part, la tesis proposa la utilització de l'aprenentatge per reforç per aprendre l'estructura interna dels comportaments. Aquest tipus d'aprenentatge és adequat per entorns desconeguts i el procés d'aprenentatge es realitza al mateix temps que el robot està explorant l'entorn. La tesis presenta també un estat de l'art d'aquest camp, en el que es detallen els principals problemes que apareixen en utilitzar els algoritmes d'aprenentatge per reforç en aplicacions reals, com la robòtica. El problema de la generalització és un dels que més influeix i consisteix en permetre l'ús de variables continues sense augmentar substancialment el temps de convergència. Després de descriure breument les principals metodologies per generalitzar, la tesis proposa l'ús d'una xarxa neural combinada amb l'algoritme d'aprenentatge per reforç Q_learning. Aquesta combinació proporciona una gran capacitat de generalització i una molt bona disposició per aprendre en tasques de robòtica amb exigències de temps real. No obstant, les xarxes neurals són aproximadors de funcions no-locals, el que significa que en treballar amb un conjunt de dades no homogeni es produeix una interferència: aprendre en un subconjunt de l'espai significa desaprendre en la resta de l'espai. El problema de la interferència afecta de manera directa en robòtica, ja que l'exploració de l'espai es realitza sempre localment. L'algoritme proposat en la tesi té en compte aquest problema i manté una base de dades representativa de totes les zones explorades. Així doncs, totes les mostres de la base de dades s'utilitzen per actualitzar la xarxa neural, i per tant, l'aprenentatge és homogeni. Finalment, la tesi presenta els resultats obtinguts amb la arquitectura de control basada en comportaments i l'algoritme d'aprenentatge per reforç. Els experiments es realitzen amb el robot URIS, desenvolupat a la Universitat de Girona, i el comportament après és el seguiment d'un objecte mitjançant visió per computador. La tesi detalla tots els dispositius desenvolupats pels experiments així com les característiques del propi robot submarí. Els resultats obtinguts demostren la idoneïtat de les propostes en permetre l'aprenentatge del comportament en temps real. En un segon apartat de resultats es demostra la capacitat de generalització de l'algoritme d'aprenentatge mitjançant el "benchmark" del "cotxe i la muntanya". Els resultats obtinguts en aquest problema milloren els resultats d'altres metodologies, demostrant la millor capacitat de generalització de les xarxes neurals.

A proposal to estimate the motion of an underwater vehicle through visual mosaicking

This thesis proposes a solution to the problem of estimating the motion of an Unmanned Underwater Vehicle (UUV). Our approach is based on the integration of the incremental measurements which are provided by a vision system. When the vehicle is close to the underwater terrain, it constructs a visual map (so called "mosaic") of the area where the mission takes place while, at the same time, it localizes itself on this map, following the Concurrent Mapping and Localization strategy. The proposed methodology to achieve this goal is based on a feature-based mosaicking algorithm. A down-looking camera is attached to the underwater vehicle. As the vehicle moves, a sequence of images of the sea-floor is acquired by the camera. For every image of the sequence, a set of characteristic features is detected by means of a corner detector. Then, their correspondences are found in the next image of the sequence. Solving the correspondence problem in an accurate and reliable way is a difficult task in computer vision. We consider different alternatives to solve this problem by introducing a detailed analysis of the textural characteristics of the image. This is done in two phases: first comparing different texture operators individually, and next selecting those that best characterize the point/matching pair and using them together to obtain a more robust characterization. Various alternatives are also studied to merge the information provided by the individual texture operators. Finally, the best approach in terms of robustness and efficiency is proposed. After the correspondences have been solved, for every pair of consecutive images we obtain a list of image features in the first image and their matchings in the next frame. Our aim is now to recover the apparent motion of the camera from these features. Although an accurate texture analysis is devoted to the matching pro-cedure, some false matches (known as outliers) could still appear among the right correspon-dences. For this reason, a robust estimation technique is used to estimate the planar transformation (homography) which explains the dominant motion of the image. Next, this homography is used to warp the processed image to the common mosaic frame, constructing a composite image formed by every frame of the sequence. With the aim of estimating the position of the vehicle as the mosaic is being constructed, the 3D motion of the vehicle can be computed from the measurements obtained by a sonar altimeter and the incremental motion computed from the homography. Unfortunately, as the mosaic increases in size, image local alignment errors increase the inaccuracies associated to the position of the vehicle. Occasionally, the trajectory described by the vehicle may cross over itself. In this situation new information is available, and the system can readjust the position estimates. Our proposal consists not only in localizing the vehicle, but also in readjusting the trajectory described by the vehicle when crossover information is obtained. This is achieved by implementing an Augmented State Kalman Filter (ASKF). Kalman filtering appears as an adequate framework to deal with position estimates and their associated covariances. Finally, some experimental results are shown. A laboratory setup has been used to analyze and evaluate the accuracy of the mosaicking system. This setup enables a quantitative measurement of the accumulated errors of the mosaics created in the lab. Then, the results obtained from real sea trials using the URIS underwater vehicle are shown.

Hybrid coordination of reinforcement learning-based behaviors for AUV control

This paper proposes a hybrid coordination method for behavior-based control architectures. The hybrid method takes advantages of the robustness and modularity in competitive approaches as well as optimized trajectories in cooperative ones. This paper shows the feasibility of applying this hybrid method with a 3D-navigation to an autonomous underwater vehicle (AUV). The behaviors are learnt online by means of reinforcement learning. A continuous Q-learning implemented with a feed-forward neural network is employed. Realistic simulations were carried out. The results obtained show the good performance of the hybrid method on behavior coordination as well as the convergence of the behaviors

Augmented state Kalman filtering for AUV navigation

Addresses the problem of estimating the motion of an autonomous underwater vehicle (AUV), while it constructs a visual map ("mosaic" image) of the ocean floor. The vehicle is equipped with a down-looking camera which is used to compute its motion with respect to the seafloor. As the mosaic increases in size, a systematic bias is introduced in the alignment of the images which form the mosaic. Therefore, this accumulative error produces a drift in the estimation of the position of the vehicle. When the arbitrary trajectory of the AUV crosses over itself, it is possible to reduce this propagation of image alignment errors within the mosaic. A Kalman filter with augmented state is proposed to optimally estimate both the visual map and the vehicle position