SiROP
Login   
Language
  • English
    • English
    • German
Home
Menu
  • Login
  • Register
  • Search Opportunity
  • Search Organization
  • Create project alert
Information
  • About SiROP
  • Team
  • Network
  • Partners
  • Advisory Board
  • Media
  • Contact
  • FAQ
  • Imprint
  • Terms & conditions
Register now After registration you will be able to apply for this opportunity online.

Learning-Guided MPC Flight

Train a neural network to predict and intermediate representation that can be used by an MPC.

  • Model predictive control (MPC) is a versatile optimization-based control method that allows to incorporate constraints directly into the control problem. The advantages of MPC can be seen in its ability to accurately control dynamical systems that include large time delays and high-order dynamics. Recent advances in compute hardware allow to run MPC even on compute-constrained quadrotors. While model-predictive control can deal with complex systems and constraints, it still assumes the existence of a reference trajectory. With this project we aim to guide the MPC to a feasible reference trajectory by using a neural network that directly predicts from camera images an expressive intermediate representation. Such tight coupling of perception and control would allow to push the speed limits of autonomous flight through cluttered environments. Requirements: - Machine learning experience (TensorFlow and/or PyTorch) - Experience in MPC preferable but not strictly required - Programming experience in C++ and Python

    Model predictive control (MPC) is a versatile optimization-based control method that allows to incorporate constraints directly into the control problem. The advantages of MPC can be seen in its ability to accurately control dynamical systems that include large time delays and high-order dynamics. Recent advances in compute hardware allow to run MPC even on compute-constrained quadrotors.
    While model-predictive control can deal with complex systems and constraints, it still assumes the existence of a reference trajectory.
    With this project we aim to guide the MPC to a feasible reference trajectory by using a neural network that directly predicts from camera images an expressive intermediate representation. Such tight coupling of perception and control would allow to push the speed limits of autonomous flight through cluttered environments.
    Requirements:
    - Machine learning experience (TensorFlow and/or PyTorch)
    - Experience in MPC preferable but not strictly required
    - Programming experience in C++ and Python

  • Evaluate different intermediate representations for autonomous flight. Implement the learned perception system in simulation and integrate the predictions into an existing MPC pipeline. If possible, deploy on a real system.

    Evaluate different intermediate representations for autonomous flight. Implement the learned perception system in simulation and integrate the predictions into an existing MPC pipeline. If possible, deploy on a real system.

  • Elia Kaufmann (ekaufmann@ifi.uzh.ch) Philipp Föhn (foehn@ifi.uzh.ch)

    Elia Kaufmann (ekaufmann@ifi.uzh.ch)
    Philipp Föhn (foehn@ifi.uzh.ch)

Calendar

Earliest start2020-10-01
Latest endNo date

Location

Robotics and Perception (UZH)

Labels

Master Thesis

Topics

  • Engineering and Technology
SiROP PARTNER INSTITUTIONS