Milan Jelisavcic

About

Milan Jelisavcic is a robotics researcher whose work sits at the exciting intersection of evolutionary computation, embodied AI, and modular robotics. His research focuses primarily on evolutionary robotics — specifically systems where both robot bodies (morphologies) and brains (controllers) co-evolve — pushing the boundaries of what autonomous machines can learn and become without direct human design. Jelisavcic's most influential contribution, "Real-World Evolution of Robot Morphologies: A Proof of Concept" (2017, 58 citations), demonstrated that physical robots could genuinely reproduce and evolve in hardware, a landmark step beyond purely simulated evolution. His sustained investigation of Lamarckian evolution — where lifetime learning acquired by an individual robot can influence its offspring — has been particularly impactful, explored across multiple papers and accumulating over 40 additional citations. He also developed Revolve, a versatile open-source simulator (30 citations) that has become a practical tool for the broader evolutionary robotics community. Across his body of work, Jelisavcic consistently tackles a fundamental challenge: how newborn robots with novel, untested bodies can rapidly acquire effective locomotion controllers. His research combining reinforcement learning, CPPNs, and CPGs offers principled solutions to this problem, making his work essential reading for anyone studying adaptive, self-designing robotic systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Real-World Evolution of Robot Morphologies: A Proof of Concept

   58 citations · 2017
2. 2
   Lamarckian Evolution of Simulated Modular Robots

   33 citations · 2019
3. 3
   Revolve: A Versatile Simulator for Online Robot Evolution

   30 citations · 2018
4. 4
   Directed Locomotion for Modular Robots with Evolvable Morphologies

   24 citations · 2018
5. 5
   Analysis of Lamarckian evolution in morphologically evolving robots

   9 citations · 2017
6. 6
   Improving RL power for on-line evolution of gaits in modular robots

   8 citations · 2016
7. 7
   Morphological Attractors in Darwinian and Lamarckian Evolutionary Robot Systems

   7 citations · 2018
8. 8
   Analysing the Relative Importance of Robot Brains and Bodies

   4 citations · 2018
9. 9
   Acquiring moving skills in robots with evolvable morphologies

   3 citations · 2017
10. 10
    Benefits of lamarckian evolution for morphologically evolving robots

    2 citations · 2017

Key Collaborators

Contact & Links