I am currently doing an internship at Motional in Singapore. My interests lie in safe and reliable machine learning and autonomous systems, applied mathematics, and compositional theories.

Before joining Motional I got my MSc at ETH Zürich focusing on robotics, machine learning, statistics, and applied category theory. My thesis was on Compositional Computational Systems. At ETH, I was working closely with Prof. Emilio Frazzoli's group and my studies were generously funded by the Excellence Scholarship & Opportunity Programme (ESOP).

Before that, I received my BSc in Aerospace Engineering from TU Delft and did my final project under the supervision of Christophe de Wagter of the Micro Air Vehicle Laboratory. I also had the pleasure to work with Ron Noomen on spacecraft trajectory optimization.

May 2021

I am happy to share that I am starting an internship at Motional in Singapore!
I will be working on detecting the safe operational envelope of the Motional self-driving stack.

October 2020

I presented the final results of my Master Thesis!
The title is *Compositional Computational Systems* and it deals with the problem of problem-solving.
You can find the presentation and the thesis itself under Publications.

July 2020

Our paper, *Integrated Benchmarking and Design for Reproducible and Accessible Evaluation of Robotic Agents*, a collaboration between ETH Zürich, Université de Montréal, and Toyota Technological Institute at Chicago, got accepted at IROS2020.

March 2020

I started my Master thesis under the supervision of Gioele Zardini and Dr. Andrea Censi in Prof. Emilio Frazzoli's group, at the Institute for Dynamic Systems and Control, ETH Zürich.
We are working on a Compositional Computational Theory in the context of Applied Category Theory.

February 2020

I concluded my semester project with Daedalean under the supervision of Prof. Thomas Hofmann.
We worked on machine learning models which utilize additional inputs at training time, showed how they can be formulated as an optimization problem over mutual information terms, and explored their implications to model certification.

January 2020

Our paper, *Learning Camera Miscalibration Detection* on which I worked at the Autonomous Systems Lab got accepted at ICRA2020.

*Compositional Computational Systems*

__Aleksandar Petrov__, supervised by Gioele Zardini, Andrea Censi, Emilio Frazzoli

Master thesis

This thesis furthermore provides the universal conditions for defining any compositional computational system. We argue that any problem can be represented as a function from the product of hom-sets of two semicategories to a rig (a kinded function) and that any procedure can also be represented as a similar kinded function. Combining all problems and procedures defined over the same subcategory of SemiCat via a solution judgment map results in a heteromorphic twisted arrow category called Laputa, which automatically provides problem-reducing and problem-solving properties.

The thesis illustrates the practical application of the theory of compositional computations systems by studying the representation of co-design problems from the theory of mathematical co-design as part of several different compositional computations systems. In the process, new results on the conditions for the solvability of co-design problems and their compositional category-theoretical properties are also presented.

*Integrated Benchmarking and Design for Reproducible and Accessible Evaluation of Robotic Agents*

Jacopo Tani, Andrea F. Daniele, Gianmarco Bernasconi, Amaury Camus, __Aleksandar Petrov__, Anthony Courchesne, Bhairav Mehta, Rohit Suri, Tomasz Zaluska, Matthew R. Walter, Emilio Frazzoli, Liam Paull, Andrea Censi

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2020

*Learning Camera Miscalibration Detection*

Andrei Cramariuc*, __Aleksandar Petrov*__, Rohit Suri, Mayank Mittal, Roland Siegwart, Cesar Cadena

IEEE International Conference on Robotics and Automation (ICRA) 2020

*Optimizing multi-rendezvous spacecraft trajectories: ΔV matrices and sequence selection*

__Aleksandar Petrov__, Ron Noomen