Course in Anomaly Detection in Autonomous Driving
Online 300-hour course in anomaly detection in autonomous driving. 6 modules covering ISO 26262 (functional safety of road vehicles), ISO 21448 (SOTIF, safety of the intended functionality) and Python (scikit-learn). Dedicated tutor and open enrolment.
Take the controls.
Run a real engineering trade-off from this field. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Formulate anomaly detection in autonomous driving as a technical problem.
Prepare sensor data and extract relevant features.
Apply statistical and machine learning detection methods.
Validate models in simulation and in operation.
Relate anomaly detection to functional safety and SOTIF.
Analyse the regulatory and insurance implications of use.
Develop an integrative anomaly detection project.
Who it's for
- Autonomous driving data engineer who wants to specialise in Anomaly Detection in Autonomous Driving.
- Anomaly detection data scientist who wants to specialise in Anomaly Detection in Autonomous Driving.
- ADAS validation engineer who wants to specialise in Anomaly Detection in Autonomous Driving.
- Functional safety specialist who wants to specialise in Anomaly Detection in Autonomous Driving.
- Autonomous mobility risk analyst who wants to specialise in Anomaly Detection in Autonomous Driving.
- Machine learning engineer who wants to specialise in Anomaly Detection in Autonomous Driving.
- Graduates in engineering or science seeking to enter the automotive industry.
What you'll work with
- ISO 26262 (functional safety of road vehicles)
- ISO 21448 (SOTIF, safety of the intended functionality)
- ISO/PAS 8800 (artificial intelligence in road vehicles)
- UL 4600 (safety of autonomous products)
- ISO/IEC 23894 (AI risk management)
- Regulation (EU) 2024/1689 (AI Act)
- UNECE Regulation No. 157 (ALKS)
6 modules · 300 h
1.1What an anomaly is in automated driving1.2Sensor failure versus unusual situation1.3Edge cases and their nature1.4Progressive performance degradation1.5Difference from object detection1.6Consequences of failing to detect in time1.7Domain vocabulary and acronyms1.8Available data sources1.9Realistic expectations of the technique1.10Exercise: classifying types of anomaly
2.1Sensor and actuator data2.2Decision system data2.3Logging of driver interventions2.4Context and environment data2.5Sampling frequency and synchronisation2.6Data quality and missing values2.7Labelling anomalies and its difficulty2.8Class imbalance2.9Privacy and minimisation2.10Practical: preparing a dataset
3.1Physical rules and thresholds3.2Classical statistical methods3.3Unsupervised detection3.4Autoencoders and reconstruction3.5Time series models3.6Distribution drift detection3.7Combining methods3.8Threshold tuning and error trade-off3.9Explainability of alerts3.10Practical: building a detector
4.1Detection metrics and their interpretation4.2False positives and their operational cost4.3Validation with real and simulated data4.4Testing in critical scenarios4.5On-board and cloud deployment4.6Latency and available resources4.7Continuous monitoring of the detector4.8Retraining and version management4.9Alert management and escalation4.10Practical: detector validation plan
5.1Relationship with the system safety analysis5.2Functional safety requirements5.3Applicable artificial intelligence regulation5.4Event logging and traceability5.5Use of alerts in maintenance5.6Use by insurers and its limits5.7Driver data protection5.8Transparency towards the user5.9Model governance5.10Practical: alert usage policy
6.1Choosing the system and type of anomaly6.2Dataset design6.3Building the detector6.4Validation plan and metrics6.5Operation and alert design6.6Regulatory and privacy analysis6.7Governance and retraining6.8Monitoring indicators6.9Methodological report6.10Project defence
How you'll learn
100% online training on the SEIUM Institute campus, with downloadable material, recorded classes and an assigned technical tutor. No fixed timetables: learners progress at their own pace within the intake, which makes it possible to combine the course with work.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is Python (scikit-learn), PyTorch, ROS 2, CARLA, MATLAB, in free, academic or evaluation versions. Tests are studied using real anomaly detection procedures on sensor series, model validation in simulation, and analysis of driving edge cases. Assessment is continuous, module by module, plus the final integrative project.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Formulate anomaly detection in autonomous driving as a technical problem.
Prepare sensor data and extract relevant features.
Apply statistical and machine learning detection methods.
Validate the models in simulation and in operation.
Relate anomaly detection to functional safety and SOTIF.
Where this takes you
Requirements
- No prior degree is required.
- Internet connection and a computer capable of running the programme's software.
- Sufficient Spanish to read technical regulations.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme comprises 300 hours in total, spread over 3 to 5 months. That is an average of 10 to 15 hours a week, compatible with a full-time job.
The course uses Python (scikit-learn), PyTorch, ROS 2, CARLA and MATLAB. Free, academic or evaluation versions are used, and the tutor indicates the free alternative where one exists.
An in-house diploma from SEIUM Institute upon passing the continuous assessment and the integrative project. This is in-house training, not an official Ministry qualification.
Yes. Modules passed in a course or postgraduate diploma in the same area are recognised towards the advanced programme. Please check the recognition policy when you enrol.
Related programmes
Course in Digital Compliance Implementation in Insurance
Online course of 300 hours in digital compliance implementation in insurance. 6 modules covering ISO 37301 (compliance management systems), Regulation (EU) 2022/2554 (DORA) and OneTrust. Assigned…
Course in Blockchain Applied to Insurance
Online 300-hour course in blockchain applied to insurance. 6 modules covering Regulation (EU) 2023/1114 (MiCA), Regulation (EU) 910/2014 (eIDAS) and Hyperledger Fabric. Assigned tutor and open…
Course in Actuarial Scenario Simulation
Online 300-hour course in actuarial scenario simulation. 6 modules covering Directiva 2009/138/CE (Solvency II, art. 45 ORSA), Delegated Regulation (EU) 2015/35 and Moody’s Scenario Generator. Assigned…
More in Insurance, Risk & Actuarial Data
All Insurance, Risk & Actuarial Data programmes → · How to apply · Fees & financing · Scholarships