Simulation-based validation of lane-keeping system

Raghoji, Sumedh Sanjay

Permalink

https://hdl.handle.net/2142/122069 Copy

Description

Title

Simulation-based validation of lane-keeping system

Author(s)

Raghoji, Sumedh Sanjay

Issue Date

2023-12-07

Director of Research (if dissertation) or Advisor (if thesis)

• Mitra, Sayan
• Ornik, Melkior

Department of Study

Aerospace Engineering

Discipline

Aerospace Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• simulation
• Validation
• lane-keeping system
• Stanley controller

Abstract

Simulation plays an important role in validating the safety of autonomous vehicles. Simulators reduce the time and cost required for road tests. In this thesis, we focus on validating the lane-keeping systems based on simulations. This validation framework contains code for the lane-keeping system and its analysis. It uses a popular sliding window approach to recognize the lanes and a Stanley controller to track the path. This simulator is based on ROS which allows for the development of modular pipelines. This pipeline makes the source code easy to modify and use. By using the provided performance metrics in the validation framework, user can evaluate their modified code against previous code. The simulator used for this validation framework is Gazebo which allows the use of distinct vehicle models, tracks, obstacles, and environments. In the validation framework, parameters like tracks, initial conditions, and velocity are varied. Gaussian noise is added to mimic the sensor noise. These parameters help in simulating real-world scenarios. The effect of varying these parameters on the prediction of cross-track and heading errors is checked against actual cross-track and heading errors. This validation framework can be used by a user to test the performance of modified lane-keeping system code easily and quickly. It provides users with the flexibility to change the parameters of the performance metrics.

Graduation Semester

2023-12

Type of Resource

Thesis

Copyright and License Information

Copyright 2023 Sumedh Sanjay Raghoji

Owning Collections