Intelligent Vehicles ME

ME41106 Intelligent Vehicles is an introduction course on the (on-board) AI technology of automated driving. Whereas the application setting is that of vehicles, most concepts apply to the broader mobile robotics setting. Students seeking a one-course overview of AI for mobile robotics are thus encouraged to attend, from Mechanical Engineering but also from outside (e.g. EEMCS, Civil Engineering and Aerospace Engineering faculties).

Introduction

Course organization:

• Intelligent Vehicles (IV) domain

• Motivation for automated driving

• SAE levels of automation

• Main components of IV technology

• Remaining challenges

• Sensor Overview (camera, radar, LiDAR)

• 3D Machine vision

• Perspective camera model

• Extrinsic and intrinsic camera transformations

• Stereo vision

• Object Detection and Classification

• Detection vs. classification

• Object proposals

• Handcrafted features (e.g. HOG) en classification (e.g. linear SVM)

• End-to-end learning: neural networks, deep learning

• Performance metrics: confusion matrices, precision vs. recall, ROC curves

• State Estimation

• Bayesian Filtering

• Kalman Filtering

• Particle Filtering

• Object Tracking

• Data Association

• Track Management

• Self-Localization en Sensor Fusion

• Absolute vs. relative localization

• Ego-motion compensation (e.g. odometry)

• Extrinsic sensor calibration

• Environment representations (grids, voxels)

• Motion Planning

• Planning as graph-search (the A* algorithm)

• Trajectory generation and optimization

• The Road Ahead

• IV technology currently on the market

• Future developments

  The course has a substantial practicum component (about 65% of course work load), where learned concepts are put into practice by means of programming assignments (Python, using Jupyter notebooks).

Final practicum assignment (50%)
- Individual assignment
- Late submission: max. one day late for -1.0 grade point
- Students having an initial grade of < 5.8 will be offered an opportunity for a remake to obtain a (maximum) grade of 5.8 for this component in the Q3 quarter.

Final written exam (50%)
- Written exam on campus.

Knock-out criteria
- At most one of the Pass/Fail practicum assignments is Fail
- The minimum final practicum assignment grade is 5
- The minimum final written exam grade is 5
- The minimum final course grade is 6

The final practicum assignment will be evaluated on the computers in the designated lab rooms. If a student chooses to work using own computing equipment, the student needs to test the software on the designated computers before handing-in.

Both final practicum assignment and final written exam results remain valid for one year. Each result can separately be transferred to the next academic year, upon request.