Post-doctoral position in model predictive control for autonomous construction machines

6 May 2024
Job Information

Organisation/Company

IRSEEM

Department

Seine Maritime

Research Field

Engineering » Control engineering

Researcher Profile

Recognised Researcher (R2)

Country

France

Application Deadline

1 Jul 2024 - 00:00 (Europe/Paris)

Type of Contract

Temporary

Job Status

Full-time

Hours Per Week

35

Offer Starting Date

6 May 2024

Is the job funded through the EU Research Framework Programme?

Not funded by an EU programme

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

No

Offer Description

Context and objectives

The offer is proposed within the frame of “PROMETE” project (Planning autonomous robots for eco-friendly construction sites) supported by the “Projets collaboratifs I-Démo régionalisé” program and co-funded by the Region Normandie, the French government (France 2030) and the European Union (FEDER). This project is carried out by a consortium of four academic and industrial partners including Heracles Robotics (https://www.heracles-robotics.com/ ) as a leader. It aims to automate construction machines to increase their efficiency and reduce their carbon footprint.The aim is to equip these machines with sensors enabling them to analyse their environment, plan their movements and interactions with other machines on-site, and carry out their tasks quickly and accurately. The methods developed must guarantee the safety of the machines when working alongside other machines, as well as their robustness in the face of terrain variations and ageing. As a partner in the “PROMETE” project, IRSEEM’s Control & Systems team, is tasked with proposing robust predictive control laws to ensure that planned tasks are carried out accurately and rapidly.

 

Missions

In collaboration with the project’s members, the 24-month-recruited Postdoctoral Researcher will have a number of tasks involving modelling, control, and implementation of the resulting control algorithms. These tasks are divided into three main parts, more or less sequential.

The first task involves modelling the construction machines under consideration. This modelling will be carried out using a geometric approach, in particular through the use of the notion of Lie group [1]. Indeed, a construction machine can be considered as a sequence of joints placed on a mobile robot [2], thus justifying this type of modelling. Different types of machines will have to be modelled, and the parameters will then be obtained using data from our partners in the project.

Based on these models, the second part of the mission will be to develop Lie group-based predictive control laws [3, 4, 5, 6, 7]. These control laws will be used to improve the precision, speed, and fuel consumption of the machines studied. To achieve this, an economic model predictive control approach will be used [8].

Finally, the third part of the mission will seek to make the predictive control laws previously studied robust [9, 10]. Indeed, the machines under consideration are subject to significant parametric variations as they age. Thus, the considered control strategies will have to be adapted to compensate for these parametric uncertainties [11, 12, 13].

In parallel with these tasks, the person recruited will be studying methods for tuning the parameters of predictive control laws. These methods are based on the use of reinforcement learning algorithms [14, 15]. The methods developed will be deployed on our partners’ construction machinery. Experiments on several regional sites are planned. Finally, the person recruited will contribute to the supervision of interns involved in the project.

 

References

F. Bullo and A. D. Lewis, Geometric Control of Mechanical Systems, ser. Texts in Applied Mathematics. Springer, 2004, vol. 49.

L. Sciavicco and B. Siciliano, Modelling and control of robot manipulators. Springer, 2001.

J. Park and K. Kim, “Tracking on Lie group for robot manipulators,” in 2014 11th International Conference on Ubiquitous Robots and Ambient Intelligence. IEEE, 2014, pp. 579–584.

S. Teng, D. Chen, W. Clark, and M. Ghaffari, “An error-state model predictive control on connected matrix Lie groups for legged robot control,” in 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2022, pp. 8850–8857.

U. V. Kalabić, R. Gupta, S. Di Cairano, A. M. Bloch, and I. V. Kolmanovsky, “MPC on manifolds with an application to the control of spacecraft attitude on SO(3),” Automatica, vol. 76, pp. 293–300, 2017.

J. Jang, S. Teng, and M. Ghaffari, “Convex geometric trajectory tracking using Lie algebraic MPC for autonomous marine vehicles,” arXiv, 2023.

T. Löw and S. Calinon, “Geometric algebra for optimal control with applications in manipulation tasks,” IEEE Transactions on Robotics, pp. 1–15, 2023.

L. Grüne and J. Pannek, Nonlinear Model Predictive Control, 2nd ed., ser. Communications and Control Engineering. Springer, 2017.

D. Limón, I. Alvarado, T. Alamo, and E. F. Camacho, “Robust tube-based MPC for tracking of constrained linear systems with additive disturbances,” Journal of Process Control, vol. 20, no. 3, pp. 248–260, 2010.

S. V. Raković, “The implicit rigid tube model predictive control,” Automatica, vol. 157, p. 111234, 2023.

B. Sakhdari and N. L. Azad, “Adaptive tube-based nonlinear mpc for economic autonomous cruise control of plug-in hybrid electric vehicles,” IEEE Transactions on Vehicular Technology, vol. 67, no. 12, pp. 11 390–11 401, 2018.

A. J. Prado, M. Torres-Torriti, and F. A. Cheein, “Distributed tube-based nonlinear MPC for motion control of skid-steer robots with terra-mechanical constraints,” IEEE Robotics and Automation Letters, vol. 6, no. 4, pp. 8045–8052, 2021.

C. Shi, Y. Yu, Y. Ma, and D. E. Chang, “Constrained control for systems on matrix Lie groups with uncertainties,” International Journal of Robust and Nonlinear Control, vol. 33, no. 5, pp. 3285–3311, 2022.

E. Bøhn, S. Gros, S. Moe, and T. A. Johansen, “Optimization of the model predictive control meta-parameters through reinforcement learning,” Engineering Applications of Artificial Intelligence, vol. 123, p. 106211, 2023.

W. Cai, S. Sawant, D. Reinhardt, S. Rastegarpour, and S. Gros, “A learning-based model predictive control strategy for home energy management systems,” IEEE Access, 2023.

 

 

 

 

Requirements

Research Field

Engineering » Control engineering

Education Level

PhD or equivalent

Skills/Qualifications

• Ph.D. in automatic control.

• Willingness to learn and experiment.

• Ability to present your work in English (ideally in French too), both orally and in writing.

• Open-mindedness, good interpersonal skills and ability to integrate into the existing team.

• Autonomy, curiosity, dynamism.

 

Specific Requirements

• Good knowledge of predictive control.

• Good knowledge of geometric modelling and control of robotic systems.

• Experience in Python development.

Languages

ENGLISH

Level

Good

Languages

FRENCH

Level

Good

Research Field

Engineering » Control engineering

Additional Information
Benefits

• Paid between 32 000€ and 35 000€ gross a year depending on profile and experience.
• Meal vouchers (Titres Restaurant) and transport allowance.
• Next to the campus (less than 5 minutes walking): several catering solutions, a supermarket with cultural and multimedia space, and a forest.
• The campus is accessible by public transport (bus and streetcar) and car (parking).

Eligibility criteria

Application should include a detailed CV and an accompanying letter.

Selection process

The selection is based first on CV screening and second on a personal interview (teams meeting).

Work Location(s)

Number of offers available

1

Company/Institute

IRSEEM/ESIGELEC

Country

France

State/Province

Normandie

City

Saint-Etienne-du-Rouvray

Postal Code

76800

Street

Avenue Galilee

Geofield

Where to apply

E-mail

[email protected]

Contact

State/Province

Normandie

City

Saint Etienne du Rouvray

Website

http://www.esigelec.fr

Street

Avenue Galilee

Postal Code

76800

E-Mail

[email protected]
[email protected]

STATUS: EXPIRED