Apply now
Master'sOff-Road, Special Vehicles & Machinery Applications open

Master’s in Perception in Unstructured Environments Engineering

Online master's programme of 1,900 hours in perception engineering for unstructured environments, with 12 modules covering off-road sensors, calibration, point cloud processing, deep learning, traversability analysis, person and obstacle detection, sensor fusion, data labelling, robustness and embedded deployment, with an assigned technical tutor.

Duration1900 h
ModeOnline
LanguageEnglish
Start datesSet by the faculty
AwardSEIUM Institute diploma
Enrol now View syllabus Ask an advisor
Virtual lab · try it before you enrol

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.

SIM · LIVE ●
Learning outcomes

What you will be able to do

01

Identify the perception challenges in unstructured environments.

02

Select and calibrate sensors for off-road use.

03

Process point clouds and images with deep learning.

04

Analyse terrain traversability.

05

Detect people and obstacles reliably.

06

Fuse data from multiple sensors.

07

Deploy robust perception models on embedded hardware.

Profile

Who it's for

  • Perception engineers developing systems for autonomous off-road machines.
  • Computer vision engineers working with deep learning for detection and segmentation.
  • Sensor fusion engineers combining LiDAR, radar and camera data.
  • Data engineers managing capture, labelling and datasets for autonomy.
  • Embedded software engineers deploying models on edge hardware.
  • Field robotics researchers working in agriculture, mining or construction.
  • Engineering or science graduates looking to enter the agricultural machinery sector.
Standards & tools

What you'll work with

  • ISO 18497 (safety of autonomous agricultural machines)
  • ISO 17757 (safety of autonomous machines in mining)
  • ISO 21448 (safety of the intended functionality)
  • IEC 60825-1 (safety of laser products)
  • Regulation (EU) 2016/679 (data protection)
  • Regulation (EU) 2024/1689 (artificial intelligence)
  • PyTorch
  • Open3D
  • ROS 2
  • NVIDIA TensorRT
  • CVAT
Syllabus

12 modules · 1900 h

  1. 1.1Unstructured work environments
  2. 1.2Absence of lanes and signage
  3. 1.3Irregular and changing terrain
  4. 1.4Vegetation and its ambiguity
  5. 1.5Dust, rain and snow
  6. 1.6Extreme lighting and backlighting
  7. 1.7Atypical objects in the environment
  8. 1.8People in work areas
  9. 1.9Perception requirements
  10. 1.10Case study: perception requirements for an autonomous tractor
  1. 2.1Mechanical and solid-state LiDAR
  2. 2.2High-resolution FMCW radar
  3. 2.3High dynamic range RGB cameras
  4. 2.4Long-wave thermal cameras
  5. 2.5Stereo vision for depth
  6. 2.6Event cameras
  7. 2.7Sensor performance in dust
  8. 2.8Selecting the sensor combination
  9. 2.9Sensor mounting and protection
  10. 2.10Lab: LiDAR in a dust chamber
  1. 3.1Intrinsic camera calibration
  2. 3.2Extrinsic calibration between sensors
  3. 3.3Camera-LiDAR calibration
  4. 3.4Radar calibration
  5. 3.5Time synchronisation via PTP
  6. 3.6Online calibration
  7. 3.7Calibration errors and their effects
  8. 3.8Calibration tools
  9. 3.9Calibration maintenance
  10. 3.10Practical: extrinsic sensor calibration
  1. 4.1Filtering and noise removal
  2. 4.2Ground segmentation
  3. 4.3Object clustering
  4. 4.4Point cloud voxelisation
  5. 4.5Neural networks for point clouds
  6. 4.6ICP registration
  7. 4.7Elevation maps
  8. 4.8PCL and Open3D libraries
  9. 4.9Computational performance of processing
  10. 4.10Lab: ground segmentation in a point cloud
  1. 5.1Convolutional neural networks
  2. 5.2Object detection
  3. 5.3Semantic image segmentation
  4. 5.4Transformer architectures for vision
  5. 5.5Model training
  6. 5.6Training data management
  7. 5.7Data augmentation
  8. 5.8Evaluation metrics
  9. 5.9Model deployment
  10. 5.10Practical: segmentation of traversable terrain
  1. 6.1Terrain type classification
  2. 6.2Terrain geometry
  3. 6.3Surface roughness of the terrain
  4. 6.4Slopes and steps
  5. 6.5Inference of soil properties
  6. 6.6Self-supervised learning from driving
  7. 6.7Cost maps
  8. 6.8Uncertainty in traversability
  9. 6.9Validation with field data
  10. 6.10Case study: traversability cost map
  1. 7.1Detecting people wearing work gear
  2. 7.2Animal detection
  3. 7.3Negative obstacles such as ditches
  4. 7.4Vegetation versus rigid obstacles
  5. 7.5Required detection distances
  6. 7.6False negative rates
  7. 7.7ISO 18497 requirements
  8. 7.8Detection redundancy
  9. 7.9Tests with mannequins
  10. 7.10Case study: detection of negative obstacles
  1. 8.1Early data fusion
  2. 8.2Late fusion of detections
  3. 8.3Bayesian estimation filters
  4. 8.4Object tracking
  5. 8.5Occupancy grids
  6. 8.6Bird's-eye-view representation
  7. 8.7Uncertainty management
  8. 8.8Fusion architectures
  9. 8.9Fusion performance
  10. 8.10Lab: occupancy grid
  1. 9.1Field data capture
  2. 9.2Managing large volumes
  3. 9.3Labelling of images and point clouds
  4. 9.4Labelling tools
  5. 9.5Labelling quality control
  6. 9.6Privacy of people captured
  7. 9.7Synthetic data from simulation
  8. 9.8Public off-road datasets
  9. 9.9Data improvement cycle
  10. 9.10Practical: labelling plan
  1. 10.1Effect of dust
  2. 10.2Effect of rain
  3. 10.3Effect of fog
  4. 10.4Sun glare on cameras
  5. 10.5Night-time operation with artificial lighting
  6. 10.6Dirt on sensors
  7. 10.7Sensor self-diagnosis
  8. 10.8Cleaning systems
  9. 10.9Controlled degradation of functions
  10. 10.10Case study: strategy for an obstructed sensor
  1. 11.1Embedded graphics processors
  2. 11.2Edge computing
  3. 11.3Model optimisation
  4. 11.4Inference with TensorRT
  5. 11.5End-to-end latency
  6. 11.6Energy consumption of computing
  7. 11.7Hardware robustness
  8. 11.8Hardware certification
  9. 11.9Model updates
  10. 11.10Lab: deployment on a Jetson platform
  1. 12.1Scenario definition
  2. 12.2Sensor selection
  3. 12.3System calibration
  4. 12.4Dataset
  5. 12.5Model development
  6. 12.6System sensor fusion
  7. 12.7Robustness evaluation
  8. 12.8Deployment on embedded hardware
  9. 12.9Thesis report
  10. 12.10Master's Thesis: defence before a panel
Methodology, software & labs

How you'll learn

100% online training on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. There are no fixed timetables: you progress at your own pace within the intake, so you can combine it with your job.

Each module combines theory, guided reading of the original standard and applied exercises. The working software is PyTorch, Open3D, ROS 2, NVIDIA TensorRT and CVAT, in free, academic or evaluation versions. Testing is studied through real procedures for field data capture, detection tests with mannequins and sensor evaluation in dust. Assessment is continuous by module, plus the final Master's Thesis.

Standards covered: ISO 18497, ISO 17757, ISO 21448, IEC 60825-1, Regulation (EU) 2016/679, Regulation (EU) 2024/1689.

Assigned technical tutor:every module ends with a deliverable that your tutor marks individually.
Standards-based learning:guided reading of ISO 18497, ISO 17757 and ISO 21448 applied to real cases, not as a theoretical annex.
Industry software:PyTorch, Open3D, ROS 2, NVIDIA TensorRT and CVAT, in free, academic or evaluation versions.
Master's Thesis with defence:A complete perception system for an unstructured environment, from sensor selection and dataset to embedded deployment, defended before a panel.
Flexible format:100% online, no fixed timetables, compatible with full-time work.
Capstone project

Build it end to end

Your final project follows the real engineering workflow, step by step, reviewed by your tutor.

Sensor selection and calibration
Choose a sensor set for an off-road scenario and carry out extrinsic calibration.
Point cloud and vision models
Segment the ground in point clouds and traversable terrain in images.
Detection and fusion
Detect negative obstacles and build an occupancy grid from fused sensors.
Robustness and deployment
Define a strategy for obstructed sensors and deploy a model on a Jetson platform.
Master's Thesis
A complete perception system evaluated for robustness, defended before a panel.
Career opportunities

Where this takes you

Perception engineerComputer vision engineerSensor fusion engineerAutonomy data engineerEmbedded software engineerField robotics researcher
Admissions & scholarships

Requirements

  • A university degree in engineering, science or architecture, or equivalent demonstrable professional experience.
  • An internet connection and a computer capable of running the programme software.
  • English sufficient to read technical standards.

Ready to start?

Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.

Apply now Scholarships
FAQ

Frequently asked questions

The programme totals 1,900 hours over 12–14 months. You set your own pace within each intake, and your tutor helps you plan the workload around your job.

You work with PyTorch, Open3D, ROS 2, NVIDIA TensorRT and CVAT, using free, academic or evaluation versions. Your tutor indicates the free alternative wherever one exists.

A SEIUM Institute diploma, awarded on passing the continuous assessment and the Master's Thesis. It is SEIUM's own qualification, not an official state degree; its recognition depends on the programme completed and the regulations of each country.

Yes. Modules passed in a course or postgraduate diploma in the same area are recognised in the higher programme. Ask about the recognition policy when you enrol.

A degree in engineering, science or architecture (or equivalent professional experience), a computer able to run the programme software and English sufficient to read technical standards.

Start dates are set by SEIUM's faculty for each intake. Our admissions team will confirm the next available date when you apply.

Keep exploring

Related programmes

See all in Off-Road, Special Vehicles & Machinery
Master's8,850 $

Master’s in Equipment and Site Digital Twins Engineering

Online master's programme of 1,900 hours in digital twin engineering for equipment and sites. 12 modules covering digital twin concepts, physical modelling, data for the twin,…

1900 hOnlineOff-Road, Special Vehicles & Machinery
Master's8,850 $

Master’s in Off-Highway Trajectory Planning Engineering

Online master's programme of 1,900 hours in trajectory planning engineering for off-highway machinery, with 12 modules covering kinematic and dynamic models, environment representation, graph search, sampling-based…

1900 hOnlineOff-Road, Special Vehicles & Machinery
Master's9,970 $

Master’s in Racing Tubular Chassis and Roll Cage Engineering

Online master's programme of 1,900 hours in competition tubular chassis and roll-cage engineering, structured in 12 modules covering tubular structures, tube materials, safety regulations, CAD design,…

1900 hOnlineOff-Road, Special Vehicles & Machinery
Master’s in Perception in Unstructured Environments EngineeringEnrol
Related programmes

More in Off-Road, Special Vehicles & Machinery

Master’s in Terramechanics and Traction on Natural Soils EngineeringView programme →Master’s in Resilient GNSS-Free Localisation EngineeringView programme →Master’s in Equipment and Site Digital Twins EngineeringView programme →Master’s in Predictive Maintenance and Condition Monitoring EngineeringView programme →Master’s in Terramechanics, Tyres & Running Gear EngineeringView programme →Master’s in Off-Highway Trajectory Planning EngineeringView programme →

All Off-Road, Special Vehicles & Machinery programmes → · How to apply · Fees & financing · Scholarships

Scroll to Top
SEIUM – Institute of Advanced Engineering
Powered by  GDPR Cookie Compliance
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.