cs.HCSep 11, 2026

Real-time Learning and Evolution in Robotic Art Installations

Authors: Sofian Audry, Stephen Kelly

Organizations: School of Media, Faculty of Communication, Universit´e du Qu´ebec `a Montr´eal · Department of Computing and Software, Faculty of Engineering, McMaster University

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

We present three robotic art installations which explore the aesthetics of adaptive behavior. Through embodied machine leaning and digital evolution, these works draw viewers into an artificial ecosystem in which open-ended novelty, trial-and-error learning, competition, and cooperation emerge in real time. Research-creation practices are examined in relation to these works, focusing on how they redefine the role of artists within a human-machine collective while examining points of convergence and divergence between artistic and engineering approaches to adaptive robotics. The systems in question use learning and evolutionary processes not as a means to optimize a specific solution, but as an aesthetic experience on its own, suggesting new modes of interdisciplinary art-science research. Finally, we discuss strategies and practices to elevate the aesthetic experience for audiences, including contexts of presentation as well as temporal and material considerations for artworks based on embodied adaptive systems.

Figures & tables

Figure 1: Vessels by Sofian Audry, Stephen Kelly and Samuel St-Aubin (2015). Photo credit: Beatriz Orviz. Video link: https://vimeo.com/758502989
Figure 2: Open-Ended Ensemble (Competitive Coevolution) by Stephen Kelly (2016). Photo credit: Caitlin Sutherland. Video link: https://vimeo.com/194200199
Figure 3: Morphosis by Sofian Audry and Rosalie D. Gagné (2024). Photo credit: Léa Martin. Video link: https://vimeo.com/1035428441
Figure 4: Vessels by Sofian Audry, Stephen Kelly and Samuel St-Aubin (2010-2024) is shown in different contexts. (a) Outdoor presentation at L’Ososphère Festival 2015 (Strasbourg, France). Photo credit: Philippe Groslier. (b) Indoor presentation at Montréal City Hall during Nuit Blanche 2016 (Montréal, Canada). Photo credit: Catherine Aboumrad.
Figure 5: Vessels custom circuit detail with microcontroller, environmental sensors, motor drivers, and RGB LEDs. Infrared diodes and receivers around the circumference allow robots to communicate with each other. Photo credit: Samuel St-Aubin.
Figure 6: Adaptation at two timescales in evolutionary reinforcement learning. On the Left, an individual agent interacts with its environment through a sensorimotor interface. It learns to predict the effect of its actions in the environment throughout its lifetime. This lifetime learning loop is how adaptation occurs in Morphosis . In vessels and Open-ended Ensemble , each agent’s situated and subjective experience in the learning environment also influences selection and variation operators in an evolutionary adaptation cycle, shown on the Right.
Figure 7: Open-Ended Ensemble (detail) by Stephen Kelly (2016). View of a robotic magnetic probe crawling along a light fixture to explore a rich world of inductive hums. Photo credit: Caitlin Sutherland.
Figure 8: Morphosis by Sofian Audry and Rosalie D. Gagné (2024). View of two robots showing details of the silicon skins and illumination. Photo credit: Léa Martin.
Figure 9: Morphosis by Sofian Audry and Rosalie D. Gagné (2024). View of the visualization monitor comparing the live learning process of the robots during behavior labeled ”Rocking without rolling” in which robots are positively rewarded for generating internal motion, and negatively rewarded for displacing horizontally on the floor. Photo credit: Musée de Joliette.
More in this topic

Robotics

Robotics studies the design and control of machines that perceive, move, and act in physical environments.

See all 751 papers

Explore similar work

CardsList

  1. Alter-Art: Exploring Embodied Artistic Creation through a Robot Avatar

    Apr 29, 2026Do Won Park, Samuele Bordini, Giorgio Grioli +2Human-Robot Interaction

    Soft Robotics for Human Cooperation and Rehabilitation, Istituto Italiano di Tecnologia, via Morego, 30, Genoa, 16163, Liguria, Italy. · Centro di Ricerca “E.Piaggio”, Universita’ di Pisa, Largo Lucio Lazzarino, 1, Pisa, 56122, Tuscany, Italy. · Dipartimento di Ingegneria dell’Informazione, Universita’ di Pisa, Via Girolamo Caruso, 16, Pisa, 56122, Tuscany, Italy.

  2. Emcar: Embodied Controller for Animating Robots

    Jun 24, 2026Carlos Gomez Cubero, Elizabeth JochumRoboticsRobot Motion Generation

    CREATE, Aalborg University, Aalborg, Denmark · CyPhy Life, Robotics and AI Lab, School of Science & Technology, IE University, Madrid, Spain · Department of Communication and Psychology, Aalborg University, Aalborg, Denmark

  3. Adaptive Human-Robot Collaborative Painting Combining Preference-Based Optimization and Dynamic Motion Primitives

    Aug 3, 2026C. Cella, M. Ristic, M. Faroni +2Robot Motion GenerationHuman-Robot Collaboration

    Politecnico di Milano