My past research topics mainly involves neural networks, data mining and evolutionary computing applied to autonomous robotic control. On this page, you will find my previous research interests as well as a list of my past publications together with their relative bibtex.


A robot is a mechatronic (Mechanical and Electronics Engineering) device, incorporating sensors, actuators and a controller. Sensors provide information about the environment surrounding the robot. Actuators allow the robot to act on its environment. The controller encodes the action policy, i.e. the choice of the actuator values depending on the current state of the world, the system internal state or exterior guidance. A mobile robot is a robot able to move thanks to own batteries and motors, Autonomous robots, as opposed to tele-operated robots, are able to act autonomously. Therefore, an autonomous robot controller exclusively relies on its own internal states and sensors to select the actuator values, a.k.a. actions.

A sensor is a piece of equipment able to grab data from the environment3. Sensors provide numerical noisy partial representation of the environment. Signal processing can be applied on the raw sensor data, providing higher level information referred to as perceptions. An actuator performs a change of state of the physical device, usually being a motor. The autonomous robot relies on its controller, a.k.a. policy, defined as a function mapping the sensor values, and possibly program internal states, onto actuator values. The behaviour is defined through the actions of the robot.

My current research leads the path toward robust (auto) adaptive approaches scaling to modern robots.

Machine Learning

Indeed, computers are meant to learn. There are many applications that cannot be performed without learning. Some simple example is convincing enough: speech/writing recognition while everybody speaks or writes in a different way. The fact is that everything cannot be known, so the knowledge has to be gained from the context as an experience (Adapt the speech recognition mechanism in order to become accurate to your voice).

The Machine Learning (ML), as an important sub field of Artificial Intelligence (AI) considers the algorithms that can be efficiently applied wherever learning is required. Usually, ML focuses on difficult problems, as does AI in general. ML focuses on data mining, feature extractions, statistical learning, inference, and many more. Several resources can be found on the web, such as on the wikipedia; I especially recommend university courses available on-line.

Genetic algorithms

It is not the strongest of the species that survives, nor the most intelligent that survives. It is the one that is the most adaptable to change, C. Darwin / 1809-1882.

Hopefully, i am neither strong nor intelligent, however, the genetic algorithms are. These stochastic optimisation algorithms get inspiration from the Darwin's theory of evolution. When the solution in the research space of a problem is to difficult to find, random search, used together with heuristics, can prove efficient. The genetic algorithms works as follow:

  1. generating randomly a population of solutions in the search space
  2. evaluating the individuals of the population (in accordance to a fitness function)
  3. select probably the best individuals found in the population
  4. generate new individuals from theses selected individuals with modifications (crossover, mutation or whatever heuristic that may improve the individuals)
  5. repeat from 2 until the program is stopped or a good solution is found

Genetic algorithms, as a Meta Heuristics, can be applied to many problems such as neural network optimisation, robotic control, anticipation mechanisms, and so on.



  • [CAR10] Fall detections in humanoid walk patterns using Reservoir Computing based control architecture. R. Kanoi, C. Hartland, 5th national conference on Control Architecture of Robots, CAR'10, Douai, France, may 2010 [BibTex]
  • [CEC09] Memory-Enhanced Evolutionary Robotics: The Echo State Network Approach. C. Hartland, N. Bredeche, M. Sebag. Accepted for publication at IEEE Congress on Evolutionary Computation (IEEE CEC 2009). To appear in May 2009. [BibTex]
  • [CAP09] Robotique Evolutionnaire et Mémoire. C. Hartland, N. Bredeche, M. Sebag., Conférence francophone sur l'apprentissage automatique, CAp09 [BibTex]
  • [ROBIO07] Using Echo State Networks for Robot Navigation Behavior Acquisition, Cédric Hartland, Nicolas Bredeche, proceedings of the IEEE International conference on Robotics and Biomimetics ROBIO 2007 [BibTex]
  • [CAP07] Change Point Detection and meta-bandits for online learning in dynamic environments, Cédric Hartland, Sylvain Gelly, Nicolas Baskiotis, Olivier Teytaud, Michele Sebag, Conférence francophone sur l'apprentissage automatique, CAp07 [BibTex]
  • [RIVF07] Human Heuristics for a Team of Mobile Robots, Tijus, C., Bredeche, N., Kodratoff, Y., Felkin, M., Hartland, C., Besson, V. & Zibetti, E., (sous presse). Proceedings of the 5th International Conference on Research, Innovation and Vision for the Future (RIVF'07), Hanoï, IEEE Computer Society Press. [BibTex]
  • [ROBIO06] Evolutionary Robotics, Anticipation and the Reality Gap, Cédric Hartland, Nicolas Bredèche, Proceedings of the IEEE International conference on Robotics and Biomimetics ROBIO 2006 [BibTex]


  • [SSEP10] Reservoir Computing based smart-sensor, Rahul Kanoi, Cédric Hartland, SAB 2010 workshop on "Smarter Sensors, Easier Processing", 2010. [BibTex]
  • [NIPS06] Multi-armed Bandit, Dynamic Environments and Meta-Bandits, Cédric Hartland, Sylvain Gelly, Nicolas Baskiotis, Olivier Teytaud, Michele Sebag, Workshop On-line Trading of Exploration and Exploitation, NIPS 2006. [BibTex]
  • [ABIALS06] Evolutionary Robotics: From Simulation to the Real World using Anticipation, Cédric Hartland, Nicolas Bredèche, Workshop ABIALS 2006 [BibTex]

Ph.D. Thesis

A contribution to robust adaptive robotic control acquisition
Defended on 16 novembre 2009 having for jury: J-S. Lienard (president), C. Fonlupt, H. Paugam-Moisy, S. Doncieux, N. Bredeche (supervisor) and M. Sebag (supervisor).

  • Manuscript (in English) [pdf]
  • Defense slides (in French) [pdf]


  • Master's internship report (in French) [pdf]