Inhalt

[ 993TAMRIASV19 ] VL Introduction to Autonomous Systems

Versionsauswahl
Workload Education level Study areas Responsible person Hours per week Coordinating university
3 ECTS M1 - Master's programme 1. year (*)Artificial Intelligence Luigi Del Re 2 hpw Johannes Kepler University Linz
Detailed information
Original study plan Master's programme Artificial Intelligence 2019W
Objectives This course is intended to provide a general overview for engineers on the main aspects of automation of driving functions. While autonomous driving is a widely discussed issue at the moment, it is not a topic only for road traffic, indeed methods have been developed in different fields ranging from UAVs, to robots and of course to automotive.

The key idea of automation - in general - is to perform an action a skilled human operator could, but better/faster or more reliably. No automation removes the physical constraints, and sometimes new ones are added by the limited operation properties of the components involved, which, by the way, are also subject to errors and even failure.

Therefore, some expectations from automated functions are unrealistic, no automation system can guarantee no accidents in traffic, especially in mixed traffic conditions. Still, automation can offer enormous advantages as automated systems do not get tired, unattentive or distracted and can react much faster.

Subject Basics of vehicle dynamics

  • Classes of vehicles
  • Movement in 2D
  • Parametrization and validation of a model

Human Driving functions

  • Driving tasks
  • The driver decision process
  • The driver as a predictive control

Classification of driving automation

  • Levels of automation
  • Frozen time vs. predictive autonomous driving functions
  • Automated Driving Functions (ADFs)
  • Effect on the overall traffic

ADF at control level

  • Brake and traction limiters
  • Stability control

ADF at guidance level

  • ACC
  • Parking assistant
  • Collision risk handling
  • Other functions

ADF at navigation level

  • Ecodriving
  • Emission aware ecodriving problem

Acquisition of surrounding vehicle information

  • Sensors
  • Sensor Fusion

V2X

  • Autonomous Driving Use Cases
  • Communication Requirements
  • Cellular V2X (C-V2X)

Prediction of movement of the surrounding traffic

  • Physics-based Motion Models
  • Maneuver-based Motion Models
  • Interaction-aware Motion Models

Methods to evaluate safety

  • Current Test Concepts in the Automobile Industry
  • X-in-the-loop Simulation Methods
  • Requirements for a Test Concept
  • Safety Evaluation for Fully Automated Driving

Buildup of scenarios

  • ISO 26262

Search of the relevant cases

  • Randomized Algorithms
  • Reduction of Sample Complexity

Simulators

  • Vehicle Dynamic Simulators
  • Traffic Simulators
  • Environment Simulators
Criteria for evaluation Written exam
Methods Lecture
Language English
Study material Lecture notes and slides
Changing subject? No
On-site course
Maximum number of participants -
Assignment procedure Direct assignment