Master’s degree1,500 h · Online · 11 modules
Master's in AI Applied to Autonomous Navigation
This master's programme gives you command of artificial intelligence and machine learning techniques applied to maritime robotics. You will work on environment perception, sensor fusion, path planning and collision avoidance, with simulation and validation throughout the modules.
Duration1,500 h
ModeOnline
StartDecember 1
Modules11
Programme diagnosis
Prior knowledge check
These are the skills this programme gives you. Tick the ones you already master and we will tell you which part of the syllabus is for you.
Your diagnosis
6of 6 to cover
You are starting from scratch in this subject. The programme is designed precisely for that.
About the programme
What is the Master's in AI Applied to Autonomous Navigation?
This master's degree is your gateway to maritime robotics. It trains you in artificial intelligence, machine learning and deep learning applied to the design and development of autonomous navigation systems for vessels. The focus is on creating perception, route planning and control algorithms that allow a vessel to navigate safely and efficiently in complex environments.
The programme stands out for its real projects, applying what you learn in simulations and prototypes of autonomous navigation; its comprehensive training, from AI theory to implementation in maritime control systems; and its use of cutting-edge tools, with current software and simulation platforms.
Programme typeMaster’s degree
ModeOnline
Duration1,500 hours
Modules11
StartDecember 1
Programme highlights
- Intelligent navigation: AI algorithms for autonomous decision-making in complex maritime environments.
- Sensors and perception: use of LiDAR, radar and cameras to build accurate models of the environment.
- Simulation and testing: development and validation through advanced simulation and controlled trials.
- Regulation and ethics: the legal framework and ethical challenges of AI in maritime navigation.
- Practical projects: application in real autonomous navigation system projects.
- Advance your career in the maritime navigation of the future.
Syllabus
Syllabus of the Master's in AI Applied to Autonomous Navigation
11 modules. Select a module to see its content.
Module 1
Implementation and Optimisation of Machine Learning Algorithms in Autonomous Navigation Systems
- 1.1. Mathematical and statistical foundations of Machine Learning applied to autonomous navigation: linear algebra, differential calculus, probability and Bayesian statistics
- 1.2. Pre-processing and cleaning of sensor data: noise filtering, normalisation, balancing and anomaly detection in radar, LIDAR, camera and GNSS data
- 1.3. Design of supervised, unsupervised and reinforcement model architectures adapted to dynamic maritime environments
- 1.4. Practical implementation of classical algorithms: logistic regression, SVM, decision trees and Random Forest for obstacle detection and route classification
- 1.5. Deep neural networks and convolutional architectures (CNN) for object recognition and processing of underwater and aerial imagery
- 1.6. Reinforcement Learning models for real-time decision-making in autonomous navigation, including Q-learning and policy-based methods
- 1.7. Hyperparameter optimisation using advanced techniques: grid search, Bayesian optimisation and evolutionary algorithms to maximise system accuracy and efficiency
- 1.8. Integration of sensor fusion algorithms to improve system robustness and resilience in adverse and uncertain environments
- 1.9. Deployment of models on embedded hardware: real-time computing, energy consumption and latency considerations in autonomous navigation systems
- 1.10. Validation and performance testing: metrics specific to autonomous navigation, simulations in virtual environments and real field trials
At this stage
136 hestimated workload11 wkweeks at your paceModule 2
Advanced Integration of Smart Sensors and Data Processing in Autonomous Navigation Systems
- 2.1. Fundamentals of smart sensors: types, physical principles and operational characteristics in maritime environments
- 2.2. Architecture of multi-sensor systems: design, integration and time synchronisation for autonomous navigation
- 2.3. Advanced signal processing: adaptive filtering, data fusion and noise reduction for LIDAR, radar and sonar sensors
- 2.4. Implementation of deep neural networks for real-time classification and segmentation of sensor data
- 2.5. Dynamic calibration and sensor self-tuning algorithms to ensure accuracy in changing sea conditions
- 2.6. Multimodal sensor fusion: Bayesian techniques, extended Kalman filters and neural networks for robust state estimation
- 2.7. Integration of inertial sensors (IMU), high-precision GNSS and visual navigation systems for positioning optimisation
- 2.8. Obstacle detection and tracking models using data from heterogeneous sensors
- 2.9. Distributed processing and edge computing to minimise latency and maximise energy efficiency on autonomous marine platforms
- 2.10. Communication protocols and standards for secure and reliable data transmission between sensors and central systems
At this stage
136 hestimated workload11 wkweeks at your paceModule 3
Advanced Machine Learning and Sensor Integration Strategies for High-Precision Autonomous Navigation Systems
- 3.1. Machine Learning fundamentals for Autonomous Navigation: supervised, unsupervised and reinforcement algorithms applied to onboard systems
- 3.2. Deep Neural Networks (DNN): advanced architectures, optimisation and regularisation specific to real-time onboard processing
- 3.3. Probabilistic and Bayesian inference models for uncertainty in perception and decision-making in dynamic marine environments
- 3.4. Multimodal sensor integration: fusion of data from LIDAR, radar, GNSS, RGB-D cameras, IMU and acoustic sensors for robust perception
- 3.5. Calibration and time synchronisation algorithms for heterogeneous sensors on autonomous mobile maritime platforms
- 3.6. Convolutional neural networks (CNN) for obstacle detection and recognition, maritime signage and analysis of environmental conditions
- 3.7. Application of advanced SLAM (Simultaneous Localization and Mapping) techniques and semantic mapping for navigation in coastal and port environments
- 3.8. Trajectory optimisation based on deep reinforcement learning for safe and efficient autonomous manoeuvres in complex scenarios
- 3.9. Assessment and mitigation of sensor errors using advanced Bayesian filters: Extended Kalman Filter (EKF), Particle Filter (PF) and their combination with neural networks
- 3.10. Hybrid system architectures for integrating predictive models and reactive planning in autonomous decision-making
At this stage
136 hestimated workload11 wkweeks at your paceModule 4
Predictive Control, Trajectory Planning and Verification of Autonomous Navigation Systems (Simulation, Testing and Certification)
- 4.1. Fundamentals of predictive control in autonomous systems: Model Predictive Control (MPC), objective functions and constraints in maritime navigation
- 4.2. Dynamic modelling of autonomous vehicles: mathematical representation of maritime systems, kinematic and dynamic modelling, and adaptation to variable environments
- 4.3. Optimal trajectory planning: search algorithms, heuristic search, continuous optimisation and their application to autonomous maritime routes
- 4.4. Sensor integration for planning: fusion of LIDAR, radar, GNSS and camera data for accurate generation of maps and obstacles
- 4.5. Resilient autonomous navigation systems: handling uncertainty, fault detection and design of redundant strategies
- 4.6. Advanced simulation for system validation: real-time simulation tools, digital twins and virtual environments for control and planning tests
- 4.7. Testing and verification methodologies: functional testing, formal validation and stress testing under real and extreme conditions
- 4.8. Certification of autonomous navigation systems: international regulations, technical standards and audit protocols for the approval of autonomous control systems
- 4.9. Implementation of predictive control on embedded hardware: computational optimisation, latency and robust design for critical marine environments
- 4.10. Professional practice and case studies: detailed analysis of real implementations, lessons learned and best practices in predictive control and trajectory planning
At this stage
136 hestimated workload11 wkweeks at your paceModule 5
Design and Architecture of Deep Neural Networks for Control in Intelligent Autonomous Navigation
- 5.1. Fundamentals and types of deep neural networks: multilayer perceptrons, convolutional networks (CNN), recurrent networks (RNN), LSTM and Transformers applied to navigation systems
- 5.2. Advanced mathematical theory for neural architecture design: activation functions, non-convex optimisation, regularisation, backpropagation and stochastic gradient descent
- 5.3. Design and parameterisation of control agents based on Deep Reinforcement Learning for decision-making in dynamic autonomous navigation environments
- 5.4. Multi-sensor integration in neural networks: processing and fusion of data from LiDAR, radar, cameras, GNSS, IMU and sonar for robust environmental perception
- 5.5. Hybrid architectures for predictive control: combining neural networks with physical models and classical controllers to ensure safety and operational efficiency
- 5.6. Advanced training and validation techniques: handling synthetic and real datasets, data augmentation techniques, class balancing and strategies to avoid overfitting in maritime contexts
- 5.7. Implementation of automatic attention systems and interpretation mechanisms (explainability) to explain neural network decisions in autonomous navigation
- 5.8. Computational and hardware optimisation: implementation on GPUs, TPUs and FPGAs to ensure real-time inference and low latency in onboard systems
- 5.9. Robustness and fault tolerance: error detection and mitigation techniques, adversarial learning and testing under extreme conditions (adverse weather, electromagnetic interference)
- 5.10. Case studies and industrial frameworks for implementing deep neural networks on autonomous navigation platforms: from simulation to open-sea validation
At this stage
136 hestimated workload11 wkweeks at your paceModule 6
Innovation in Machine Learning Algorithms, Sensor Integration and Predictive Control for High-Precision Autonomous Navigation Systems
- 6.1. Advanced fundamentals of machine learning applied to autonomous navigation: supervised, unsupervised and reinforcement learning, with emphasis on adaptation in dynamic maritime environments
- 6.2. Design and innovation in environmental perception algorithms: convolutional neural networks (CNN), recurrent networks (RNN) and Transformers for real-time sensor processing
- 6.3. Advanced multi-sensor integration: fusion of data from LIDAR, radar, hyperspectral cameras, GNSS systems and inertial sensors to build accurate digital models
- 6.4. Model predictive control (MPC) for safe manoeuvres, route optimisation and adaptive response to hydrodynamic and meteorological disturbances
- 6.5. Design of hybrid control architectures: combining classical PID control with deep learning techniques to improve stability and energy efficiency
- 6.6. Implementation of collision detection and avoidance systems using trajectory planning algorithms based on Deep Reinforcement Learning and genetic algorithms
- 6.7. Large-scale model simulation and training: creating digital environments for virtual validation using digital twins and high-fidelity maritime simulators
- 6.8. Real-time optimisation and parameter self-tuning: adaptive mechanisms to maintain accuracy under changing operating conditions
- 6.9. Performance evaluation and key metrics in autonomous navigation: positional accuracy, computational efficiency, robustness to failures and tolerance of sensor errors
- 6.10. Technical and regulatory aspects for the safe and reliable integration of intelligent systems on maritime platforms: interoperability, standards and specialised communication protocols
At this stage
136 hestimated workload11 wkweeks at your paceModule 7
Integrated Machine Learning, Advanced Sensor Technology and Predictive Control Strategies for High-Performance Autonomous Navigation Systems
- 7.1. Advanced Machine Learning fundamentals for autonomous navigation: supervised, unsupervised and reinforcement algorithms applied to dynamic maritime environments
- 7.2. Implementation of deep neural networks (Deep Learning) for real-time obstacle recognition and classification
- 7.3. Multispectral sensor fusion: integration of LiDAR, synthetic aperture radar (SAR), hyperspectral camera and ultrasonic sensor data for environmental perception
- 7.4. Predictive modelling based on time series to anticipate wave patterns, ocean currents and adverse weather conditions
- 7.5. Design and optimisation of model predictive control (MPC) for active stabilisation and manoeuvrability of autonomous vessels in highly uncertain scenarios
- 7.6. Calibration and cross-validation protocols for specialised maritime sensors to ensure accuracy and redundancy
- 7.7. Integration of adaptive and autonomous algorithms for continuous decision-making in navigation systems with multiple heterogeneous variables
- 7.8. Robust digital simulation: creation of realistic virtual environments for training and testing AI models applied to complex maritime routes
- 7.9. Advanced embedded system architectures and edge computing for real-time processing and latency reduction on the autonomous bridge
- 7.10. Cybersecurity and operational resilience strategies on autonomous platforms against attacks and sensor or actuator failures
At this stage
136 hestimated workload11 wkweeks at your paceModule 8
Multimodal Perception and Data Fusion Systems for Autonomous Navigation in Dynamic Environments
- 8.1. Fundamentals of multimodal perception: principles, sensory modalities and challenges in dynamic environments
- 8.2. Sensor technologies for autonomous navigation: LiDAR, radar, RGB-D cameras, ultrasound and inertial sensors
- 8.3. Advanced signal processing: filtering, calibration and correction of external sensor data
- 8.4. Sensor fusion based on probabilistic models: Kalman Filters, Particle Filters and Bayesian networks for robust integration
- 8.5. Real-time mapping and generation of 3D spatial representations: SLAM (Simultaneous Localization and Mapping) and its variants
- 8.6. Object detection and classification using deep learning: CNN, RNN and Transformer architectures applied to navigation
- 8.7. Semantic segmentation and obstacle detection algorithms in complex coastal and maritime scenarios
- 8.8. Fusion of heterogeneous data: time synchronisation, spatial alignment and quality assessment of multiple sources
- 8.9. Decision-making methods based on multimodal perception for safe and efficient autonomous manoeuvres
- 8.10. Real-time implementation: embedded hardware, computational optimisation and parallelisation techniques
At this stage
136 hestimated workload11 wkweeks at your paceModule 9
Innovation and Development of Advanced Algorithms in Integrated Autonomous Navigation Systems
- 9.1. Mathematical and statistical foundations applied to autonomous navigation: linear algebra, vector calculus and probability theory
- 9.2. Design and optimisation of multi-source perception algorithms: integration of LiDAR, radar, camera and inertial sensor data
- 9.3. Deep neural networks and reinforcement learning for decision-making in dynamic marine environments
- 9.4. Modelling and simulation of kinematic and dynamic systems of autonomous maritime vehicles using non-linear systems and adaptive control
- 9.5. Implementation of advanced sensor fusion algorithms to improve accuracy in estimating vehicle and environment state
- 9.6. Development of resilient and redundant AI-based navigation systems for fault mitigation and robustness against external interference
- 9.7. Application of real-time optimal route planning algorithms considering environmental, safety and energy-efficiency constraints
- 9.8. Evaluation and validation using high-realism simulators and test benches in virtual maritime environments
- 9.9. Integration of advanced computer vision techniques for recognition, classification and tracking of moving objects and obstacles
- 9.10. International regulations, standards and protocols for implementing autonomous systems in the maritime domain: practical application and regulatory challenges
At this stage
136 hestimated workload11 wkweeks at your paceModule 10
Regulation, Ethics and Certification of Autonomous Navigation
- 10.1. IMO MASS Code
- 10.2. Levels of ship autonomy
- 10.3. COLREG and autonomous systems
- 10.4. Legal liability of AI systems
- 10.5. EU Artificial Intelligence Act
- 10.6. Certification by classification societies
- 10.7. Remote control centres
- 10.8. Human factors in supervision
- 10.9. Cybersecurity of autonomous ships
- 10.10. International pilot projects
At this stage
136 hestimated workload11 wkweeks at your paceModule 11
Master's Thesis in AI Applied to Autonomous Navigation: Development of an Integrated Perception, Planning and Control System Based on Deep Neural Networks for Autonomous Vehicles in Complex Environments
- 11.1. Definition and specifications of the final project: objectives, scope and comprehensive evaluation criteria for an integrated autonomous system.
- 11.2. Advanced perception: multimodal integration of LiDAR sensors, RGB-D cameras, radar and ultrasonic sensors for robust real-time recognition of the environment.
- 11.3. Sensor processing and fusion: algorithms for data cleaning, 3D reconstruction, semantic segmentation and pose estimation through supervised and unsupervised learning.
- 11.4. Deep neural networks: design and optimisation of CNN, RNN and Transformer architectures for contextual interpretation and detection of dynamic and static objects.
- 11.5. Development of predictive models to anticipate the movements and behaviour of human agents and vehicles in unstructured environments.
- 11.6. Trajectory planning: formulation and solution of multi-objective optimisation problems in high-dimensional spaces, using deep reinforcement learning techniques and tree-search-based algorithms.
- 11.7. Incorporation of kinematic and dynamic constraints to ensure route feasibility and safety in complex urban, maritime or aerial scenarios.
- 11.8. Adaptive and robust control: implementation of model-based predictive control algorithms, neural networks and hybrid methods for efficient stabilisation and manoeuvrability of the autonomous vehicle.
- 11.9. Advanced simulation and virtual validation: realistic simulation platforms with connected digital environments to assess performance, resilience and behaviour in edge cases.
- 11.10. System integration and architecture: modular and scalable design for the synergistic coupling of perception, planning and control, with secure communication and redundant protocols.
At this stage
136 hestimated workload11 wkweeks at your pacePlan your pace
Duration and format of the Master's in AI Applied to Autonomous Navigation
It is 1,500 hours. Tell us how many you can study each week and we will tell you when you will really finish.
4 h30 h
With 12 h per week
29
months to finish
125 weeks
weeks in total
This is the pace at which most students finish. It fits well with the instalment plan.
Faculty
Academic leadership and faculty
The Navalis teaching team: working professionals who teach and supervise the programmes.
Dra. Marina AlcoverAcademic Director and Head of Quality Assurance
Ing. Diego CifuentesSenior Lecturer in Naval Welding and Non-Destructive Testing; Head of QA/QC for Welded Joints
Cap. Javier Abaroa (MCA)Senior Lecturer in Ship Handling, Electronic Navigation (ECDIS/Radar/ARPA) and GMDSS; STCW Compliance and Bridge Resource Management Coordinator
Dr. Pau FerrerSenior Lecturer in Naval Architecture and Simulation; Director of the Hydrodynamics and CFD Laboratory
Dra. Nuria LlobregatSenior Lecturer in Marine Meteorology, SAR and Operational Safety; Navigation Risk Management Coordinator
Ing. Javier BañulsSenior Lecturer in Electrical Engineering, Automation and ETO Engineering; Electronic Integration and Engine Alarms Coordinator
Ing. Sofía MarquinaSenior Lecturer in Materials, Composites and Corrosion Protection; Coordinator of the Advanced Paints and Coatings Line
Ing. Tomás RieraSenior Lecturer in Propulsion, Alternative Energy and Hybrid Systems; Energy Performance and Vibration CoordinatorObjectives
Objectives of the Master's in AI Applied to Autonomous Navigation
01Implement advanced perception systems.
02Develop robust control algorithms.
03Optimise real-time decision-making.
04Integrate AI into complex navigation architectures.
05Ensure the reliability and safety of autonomous systems.
06Adapt navigation to dynamic and uncertain environments.
Careers
Career opportunities after the Master's in AI Applied to Autonomous Navigation
Frequently asked questions
Questions about the Master's in AI Applied to Autonomous Navigation
Land, air, maritime and space vehicles.
Yes. The pathway includes ECDIS/Radar-ARPA/BRM with port, ocean, fog, heavy weather and SAR scenarios.
Online with live sessions; hybrid option for simulator/placement periods through partnership agreements.
Autonomous navigation in all four domains: air, land, maritime and space.
Functional SMCP is recommended. We provide support materials for standard phraseology.
Yes, with a related qualification or experience in maritime/port operations. The admission interview will confirm suitability.
Optional (3–6 months) through Companies & Partnerships and the Alumni Network.
Simulator exercises (rubrics), passage plans, SOPs, checklists, micro-tests and an applied master's thesis.
Related programmes
More programmes in Bridge and navigation
Master’s degree · 1,500 hMaster's in Remote Operation of Autonomous Ships and the MASS Code$5,410 · View programme →Master’s degree · 1,500 hMaster's in Advanced Training for Masters and Bridge Officers$3,750 · View programme →Professional diploma · 800 hProfessional Diploma in Dynamic Positioning (DP) and Offshore Support Vessels$1,760 · View programme →
Online admission
Admission process
No waiting for a call back. If your documents are in order, you can enrol in the same session.
Entry requirements
- Academic or professional profile: a degree in computer engineering, mathematics, physics, robotics or a related field, or demonstrable professional experience in AI, robotics or autonomous navigation.
- Language skills: advanced technical English (B2 or higher) is recommended.
- Technical knowledge: solid skills in Python programming, linear algebra, calculus, statistics and machine learning.
- Documentation: up-to-date CV, copy of your qualification, academic transcript and a motivation letter on your interest in AI-based autonomous navigation.
- Technical requirements (online): a computer with adequate processing capacity (a GPU is recommended) and a high-speed internet connection. Enrolment is open all year round; admission is provisional until the committee validates your documentation.
- 01ApplicationYour details, background and documents.
- 02Instant reviewWe check that everything is complete and in order.
- 03EnrolmentChoose full payment or instalments and reserve your place.
Start my applicationFrom $1,636 deposit
Admission is provisional until the committee validates the documents provided. If you do not meet the requirements, the full amount is refunded.
Any questions?
Ask us before you enrol
Leave your details and admissions will write to you to answer any questions. If you are already sure, you can apply and pay in the same session.