Physics-guided learning for machine control

15 Apr 2026
Job Information

Organisation/Company

Czech Technical University in Prague

Department

Department of Cybernetics

Research Field

Computer science » Cybernetics

Researcher Profile

First Stage Researcher (R1)

Positions

PhD Positions

Application Deadline

31 May 2026 - 23:59 (Europe/Brussels)

Country

Czech Republic

Type of Contract

Temporary

Job Status

Full-time

Hours Per Week

40

Offer Starting Date

1 Sep 2026

Is the job funded through the EU Research Framework Programme?

Horizon Europe - MSCA

Is the Job related to staff position within a Research Infrastructure?

No

Offer Description

Pictus PhD Fellowship Programme

Czech Technical University in Prague – International PhD Programme (PICTUS) will recruit  a total of 23 PhD students (see Eligibility criteria below) for a full 48-month employment contract and enroll them in a specific PhD course.  Candidates should prepare their own research proposal from the topics offered  .

Application form  should be filled  and all required documents uploaded using the dedicated electronic form.
helpdesk: petra.koudelova@fsv.cvut.cz

Physics-guided learning for machine control

Description:

Robust machine control assumes modeling of robot-environment interactions. An example may include an outdoor autonomous ground robot that needs to be aware of its model and how the terrain will interact with it when a control sequence is executed. A flying robot may benefit from knowing the wind field ahead to model aerodynamic forces correctly.

However, building robust perception systems that can efficiently adapt in a self-supervised manner to novel environments remains a significant challenge. We identify three core issues: (i) black-box models that ignore the robot's physical embodiment suffer from poor generalization, weak explainability, and limited transferability; (ii) sample-inefficient learning requires large volumes of annotated, domain-specific data; and (iii) complex architectures with tightly coupled components hinder modular adaptation. To address these limitations, we research a physics-guided machine learning framework that integrates physical knowledge with data-driven methods. Physical knowledge includes, among others, kinematics and the dynamics of the robot, terrain interaction and contact models, and environmental physics, such as wind. The physics can be incorporated in various ways. Two methods now researched most intensively are i) trainable machine learning pipelines may embed differentiable physical models, and ii) the learning process may be informed by constraining the predicted variable to obey physical laws; we can see it as physics-informed losses. We will seek new ways to embed the physics.

Our approach aims to enable explainable, embodiment-aware, and probabilistically consistent adaptation from onboard sensory data via end-to-end differentiable architectures, enhancing robustness, efficiency, and generalization across diverse robotic platforms and environments.

Project 18

Physics-guided learning for machine control

Fac. of Electr. Eng.

Department of Cybernetics

https://cmp.felk.cvut.cz/~svoboda/

Short description

Robust machine control assumes modeling of robot-environment interactions. An example may include an outdoor autonomous ground robot that needs to be aware of its model and how the terrain will interact with it when a control sequence is executed. A flying robot may benefit from knowing the wind field ahead to model aerodynamic forces correctly.

However, building robust perception systems that can efficiently adapt in a self-supervised manner to novel environments remains a significant challenge. We identify three core issues: (i) black-box models that ignore the robot's physical embodiment suffer from poor generalization, weak explainability, and limited transferability; (ii) sample-inefficient learning requires large volumes of annotated, domain-specific data; and (iii) complex architectures with tightly coupled components hinder modular adaptation. To address these limitations, we research a physics-guided machine learning framework that integrates physical knowledge with data-driven methods. Physical knowledge includes, among others, kinematics and the dynamics of the robot, terrain interaction and contact models, and environmental physics, such as wind. The physics can be incorporated in various ways. Two methods now researched most intensively are i) trainable machine learning pipelines may embed differentiable physical models, and ii) the learning process may be informed by constraining the predicted variable to obey physical laws; we can see it as physics-informed losses. We will seek new ways to embed the physics.

Our approach aims to enable explainable, embodiment-aware, and probabilistically consistent adaptation from onboard sensory data via end-to-end differentiable architectures, enhancing robustness, efficiency, and generalization across diverse robotic platforms and environments.

Supervisor

 Prof. Tomáš Svoboda

(svobodat@cvut.cz ) 

 

Secondment:

R. Agishev, K. Zimmermann, V. Kubelka, M. Pecka, T. Svoboda. MonoForce: Self-supervised Learning of Physics-aware Model for Predicting Robot-terrain Interaction. IEEE-IROS 2024

V. Salansky, K. Zimmermann, T. Petricek, T. Svoboda. Pose consistency KKT-loss for weakly supervised learning of robot-terrain interaction model. IEEE Robotics and Automation Letters, 2021, Volume 6, Issue 3.

P. Vacek, D. Hurych, K. Zimmermann, and T. Svoboda. Let-It-Flow: Simultaneous Optimization of 3D Flow and Object Clustering. IEEE Transactions on Intelligent Vehicles, 2024

Z. Straka, T. Svoboda, M. Hoffmann. PreCNet: Next-Frame Video Prediction Based on Predictive Coding. IEEE Transactions on Neural Networks and Learning Systems, Vol 35, Issue 8, 2024,

P. Vacek, O. Jasek, K. Zimmermann, T. Svoboda. Learning to Predict Lidar Intensities. IEEE Transactions on Intelligent Transportation Systems. April 2022, Volume 23, Issue 4.

Where to apply

E-mail

svobodat@cvut.cz

Requirements

Research Field

Computer science » Cybernetics

Education Level

Master Degree or equivalent

Skills/Qualifications

Required documents  to be uploaded to the  Application form webpage

CV including
a)      Education received
b)      Any working and professional experience
c)      Prizes and Awards
d)      Participation at projects
e)     Internships
f)      List of publications

Master thesis
pdf max 20 MB

Research proposal
Select one of the topics offered
 
Required sections: 1) Excellence, 2) Impact incl. expected outputs of the project, 3) Quality and Efficiency of the Implementation 4), Personal motivation: description of career goals, foreseen advancement in research and professional skills of the Applicant.
Consult your proposal with the respective Supervisor. You are encouraged to communicate with the Supervisor  before you submit the application.

Master degree diploma

Specific Requirements

Experience:
* Master degree at the time of beginning of the contract
* previous studies compatible with the project they intend to apply with 

Mobility: applicants must not have resided or carried out their main activity in the Czech Republic for more than 12 months in the 3 years prior to the Calls’ deadline (career breaks not counted). 

Languages

ENGLISH

Level

Good

Additional Information
Benefits

Benefits and Conditions

Gross monthly salary of 69 300 CZK/month*
Family allowance 8 888 CZK/month (for applicants with dependent family members)
Travel support for conferences and secondments
Research costs1 

*The gross monthly salary CZK is under the standard scheme in the Czech Republic and includes mandatory social and health insurance. Therefore, the gross salary contains an employee contribution to social and health insurance of 11% and it is standardly taxable (15 % rate). Some tax discounts are given to e.g. employees with children. All rates subject to adjustments due to changing exchange rate. 

Example: a single researcher would get approx. 55 000 CZK as net salary

The offered salary is equivalent to the standard MSCA individual postdoc award, and it is highly competitive 

Eligibility criteria

Experience:
* Master degree at the time of beginning of the contract
* previous studies compatible with the project they intend to apply with 

Mobility: applicants must not have resided or carried out their main activity in the Czech Republic for more than 12 months in the 3 years prior to the Calls’ deadline (career breaks not counted). 

Website for additional job details

https://cmp.felk.cvut.cz/~svoboda/

Work Location(s)

Number of offers available

1

Company/Institute

Faculty of Electrical Engineering

Country

Czech Republic

City

Prague 2

Postal Code

120 00

Street

Karlovo náměstí 13

Geofield

Contact

State/Province

Czech Republic

City

Prague

Website

https://www.cvut.cz/en

Street

Jugoslávských partyzánů 1580/3

Postal Code

160 00

E-Mail

svobodat@cvut.cz

STATUS: EXPIRED

• X (formerly Twitter)
• Facebook
• LinkedIn
• Whatsapp
• 
  More share options
  • E-mail
  • Pocket
  • Viadeo
  • Gmail
  • Weibo
  • Blogger
  • Qzone
  • YahooMail