TUM Project Opportunities
← Back to projects

idp-217 · Informatics IDP Hub

IDP HUSKY: Autonomous Navigation for Construction Applications

Automatically extracted from the linked project document. Check the original PDF before relying on a detail.

Chair listing page ↗Original PDF ↗

Document language: English · Pages: 1 · Extracted: 25 Sept 2026, 23:29

Offer 1

IDP HUSKY: Autonomous Navigation for Construction Applications

Page 1 ↗
English summary

Students will improve and extend a ROS2-based navigation stack for the Clearpath Husky, integrating open-source SLAM and localization packages and working on path planning, smoothing, optimization, re-planning, and parameter tuning. They may integrate or implement and evaluate algorithms, with possible controller improvements or MPC adaptation. Developed components will be evaluated in simulation and, where possible, on the real Husky platform for autonomous navigation in construction environments.

Overview

Project details
IDP HUSKY: Autonomous Navigation for Construction Applications
Source excerpt · Page 1
  • “IDP HUSKY: Autonomous Navigation for Construction Applications” Page 1 ↗
Organization
Not stated in the PDF
Project formats
  • IDP
    Source excerpt · Page 1
    • “IDP HUSKY: Autonomous Navigation for Construction Applications” Page 1 ↗
Degree levels
Not stated in the PDF
Project goal
Achieve robust and efficient autonomous navigation in dynamic and unstructured construction environments.
Source excerpt · Page 1
  • “The goal is to achieve robust and efficient autonomous navigation in dynamic and unstructured construction environments.” Page 1 ↗

Topics and work

Subjects
  • Autonomous robot navigation
    Source excerpt · Page 1
    • “robust and efficient autonomous navigation” Page 1 ↗
  • Path planning, path smoothing, and optimization
    Source excerpt · Page 1
    • “path planning, path smoothing and optimization” Page 1 ↗
  • SLAM and localization
    Source excerpt · Page 1
    • “Existing open-source SLAM and localization packages will be integrated into our stack rather than developed from scratch.” Page 1 ↗
  • Re-planning
    Source excerpt · Page 1
  • Parameter tuning
    Source excerpt · Page 1
Application areas
  • Construction applications and environments
    Source excerpt · Page 1
    • “dynamic and unstructured construction environments.” Page 1 ↗
Methods and tools
  • ROS2-based navigation stack
    Source excerpt · Page 1
    • “our custom ROS2-based navigation stack” Page 1 ↗
  • Clearpath Husky mobile robot
    Source excerpt · Page 1
    • “for the Clearpath Husky mobile robot.” Page 1 ↗
  • Open-source SLAM and localization packages
    Source excerpt · Page 1
    • “Existing open-source SLAM and localization packages will be integrated into our stack rather than developed from scratch.” Page 1 ↗
  • Pursuit-based controller
    Source excerpt · Page 1
    • “include improvements to our existing pursuit-based controller” Page 1 ↗
  • Model Predictive Controller (MPC)
    Source excerpt · Page 1
    • “the adaptation of an open-source Model Predictive Controller (MPC).” Page 1 ↗
  • Simulation
    Source excerpt · Page 1
    • “The developed components will be evaluated first in simulation” Page 1 ↗
  • Real Husky platform
    Source excerpt · Page 1
Activities
  • Improve and extend the custom navigation stack.
    Source excerpt · Page 1
    • “focuses on improving and extending our custom ROS2-based navigation stack” Page 1 ↗
  • Integrate existing open-source SLAM and localization packages.
    Source excerpt · Page 1
    • “Existing open-source SLAM and localization packages will be integrated into our stack rather than developed from scratch.” Page 1 ↗
  • Work on path planning, path smoothing and optimization, re-planning, and systematic parameter tuning.
    Source excerpt · Page 1
    • “The main focus of the project is path planning, path smoothing and optimization, re-planning, and systematic parameter tuning.” Page 1 ↗
  • Integrate existing open-source solutions or implement and evaluate own algorithms.
    Source excerpt · Page 1
    • “Students may integrate existing open-source solutions or implement and evaluate their own algorithms.” Page 1 ↗
  • Potentially improve the existing pursuit-based controller or adapt an open-source MPC.
    Source excerpt · Page 1
    • “Depending on progress and interest, the project may also include improvements to our existing pursuit-based controller or the adaptation of an open-source Model Predictive Controller (MPC).” Page 1 ↗
  • Evaluate developed components in simulation and, where possible, on the real Husky platform.
    Source excerpt · Page 1
    • “The developed components will be evaluated first in simulation and, where possible, on the real Husky platform.” Page 1 ↗
Kinds of work
  • Software development
    Source excerpt · Page 1
    • “improving and extending our custom ROS2-based navigation stack” Page 1 ↗
  • Modeling and simulation
    Source excerpt · Page 1
    • “The developed components will be evaluated first in simulation” Page 1 ↗
  • Hardware and lab work
    Source excerpt · Page 1
Expected outputs
  • Developed components for the navigation stack
    Source excerpt · Page 1
    • “The developed components will be evaluated first in simulation and, where possible, on the real Husky platform.” Page 1 ↗

Requirements

Required skills
  • Proficiency in Gazebo, C++/ROS2, and Python
    Source excerpt · Page 1
    • “Proficiency in Gazebo, C++/ROS2, and Python” Page 1 ↗
  • Good understanding of robot navigation, path planning, and SLAM
    Source excerpt · Page 1
    • “Good understanding of robot navigation, path planning, and SLAM” Page 1 ↗
Recommended skills
  • Knowledge of optimization and control/MPC
    Source excerpt · Page 1
    • “Knowledge of optimization and control/MPC is a plus” Page 1 ↗
  • Previous experience with real robots
    Source excerpt · Page 1
    • “Previous experience with real robots is beneficial, though not mandatory” Page 1 ↗
Eligible study fields
Not stated in the PDF
Programming in the project
Central
Source excerpt · Page 1, Page 1
  • “improving and extending our custom ROS2-based navigation stack” Page 1 ↗
  • “Students may integrate existing open-source solutions or implement and evaluate their own algorithms.” Page 1 ↗
Programming prerequisite
Required
Source excerpt · Page 1
  • “Proficiency in Gazebo, C++/ROS2, and Python” Page 1 ↗

Practical details

Team size
Not stated in the PDF
Duration
1-2 semesters
Source excerpt · Page 1
Start
November 2026Normalized date: 2026-11
Source excerpt · Page 1
Application deadline
Not stated in the PDF
Work location
In person; specific location not stated
Source excerpt · Page 1
  • “Active (in-person) participation in the project is required.” Page 1 ↗
Location mode
On site
Source excerpt · Page 1
  • “Active (in-person) participation in the project is required.” Page 1 ↗
Working language
Not stated in the PDF

Learning and support

Learning opportunities
Not stated in the PDF
Support offered
Not stated in the PDF

Application and contacts

Contacts
How to apply
Send your CV along with a short paragraph explaining why you are a good candidate to panagiotis.petropoulakis@tum.de.
Source excerpt · Page 1
  • “Please send your CV along with a short paragraph explaining why you are a good candidate to panagiotis.petropoulakis@tum.de” Page 1 ↗
Further information
Not stated in the PDF