Student Projects

If you are interested in one of the below projects, Semester Projects, Master Projects at EPFL, or Master Projects in Industry please contact the first person of reference indicated in each description either by telephone, or by email, or by visiting us directly to the LASA offices.

Semester Projects

 

Learning Nonlinear Limit Cycles from Demonstration

Introduction

Limit cycles are frequently exploited in robotic applications where a periodic motion is desirable (e.g. polishing, walking, joint excitation, etc). A closed-loop trajectory of a dynamical system (DS) forms a stable limit cycle if all the neighbor trajectories converge to it. However, robustly controlling a robot in order to follow such a trajectory is not trivial, especially when it comes to nonlinear limit cycles. Learning the desired DS from demonstration can ensure stable tracking of complex trajectories. In this project, we are going to learn different DSs which contain a stable limit cycle with a nonlinear profile( see Figure 1).
 

Approach

From the Lyapunov stability criterion, we first propose a general form of potential function from which we can extract a dynamical system with stable limit cycles. Once the sample trajectories are recorded, we will use a nonparametric Bayesian approach (here GPR) to learn the target process of limit cycles from the demonstration.
 

Motivation

The project helps you: (a) understanding dynamical systems and the concept of learning from demonstration (b) developing and applying Gaussian Process Models in a practical example. (c) implementing the learned model on a robotic arm (KUKA IIWA 14).
 
ProjectSemester Project (LASA)
PeriodSept.21st, 2021 – Jan. 31st, 2021
Section(s)ME MT SV EL IN MA
Type50% theory, 50% software
Knowledge(s)Programming skills (Python / C++ / Matlab)
Subjects(s)Robot Planning; Control Theory;
Responsible(s)Farshad Khadivar
URLPaper; githup
AvailabilityAvailable

Incrementally Learning and Exploiting Inverse Dynamics

Introduction:

The inverse dynamics of a robotic manipulator is instrumental in precise robot control and manipulation.
However, acquiring such a model is challenging, not only due to unmodelled non-linearities such as joint friction, but also from a machine learning perspective (e.g., input space dimension, amount of data needed). The accuracy of such models, regardless of the learning techniques, relies on proper excitation and exploration of the robot’s configuration space, in order to collect a rich dataset. This study aims to Incrementally explore the robot’s workspace to provide rich data in learning the inverse dynamics of a serial robotic manipulator using supervised machine learning techniques.
 

Approach

Our learning approach, in this project, is based on exploration and exploitation. We assume that a “rigid body dynamic” model of the robot, however inaccurate, is apriori known. Then we will use this model to extract and optimize the most informative set of trajectories online. While moving along the acquired trajectory, the recorded robot states will be exploited to update the dynamic model. Since the learning rate is much slower than the robot control frequency, this scenario ought to be executed incrementally.
 

Motivation

The project helps you: (a) understanding inverse dynamics model of a robotic manipulator (b) developing and applying Bayesian optimization framework in a real robotic application. (c) implementing the learned model on a robotic arm (KUKA IIWA 14)
ProjectSemester Project (LASA)
PeriodSept.21st, 2021 – Jan. 31st, 2021
Section(s)ME MT SV EL IN MA
Type50% theory, 50% software
Knowledge(s)Programming skills (Python / C++ / Matlab)
Control Systems; ROS (a plus); Rviz (a plus)
Subjects(s)Robot Dynamics; Control Theory; Bayesian Optimization
Responsible(s)Farshad Khadivar
URLgithub
AvailabilityAvailable

Master Projects at EPFL

Swarm Furniture Obstacle Avoidance for Smart-Living Environment Assisting Wheelchair Navigation

Description:

This project is a step towards the development of a modular control architecture for collaborative robots in smart-living environments. The goal is to develop a simulation environment of this fully smart home (mobile furniture) and implement a control framework for the modular robotic system to facilitate user’s navigation and obstacle avoidance through the smart home. The project will start by analyzing and drafting a swarm robot formation that could work in decentralised control based on LASA’s developed obstacle avoidance through modulated dynamical systems [1]. The resulting controller should plan motions either to move away or towards a person while respecting the environment’s constraints and avoiding collisions.

The test-bed simulation with modular robotic furniture will operate autonomously for increasing the efficient navigation of a pedestrian user or a smart wheelchair robot user (simulated semi-autonomous driving Qolo [2]). The resulting simulator and controller will serve as a baseline for enabling further development of the human-robot interaction framework to control specific behaviours of the modular robot furniture, such as a user commanding furniture to get closer or move further away.

This project is associated with the CIS Intelligent Assistive Robotics Collaboration Research Pillar. It is co-supervised by LASA and BioRob laboratories.

Approach

  1. Develop a test-bed environment (in Unity or Rviz) containing robotic furniture that incorporates physical/computational constraints and velocity control.

  2. Implement an obstacle avoidance DS-based algorithm on each robot object acting independently [1].

  3. Analyze the behaviour of the modular swarm acting independently and evaluate the performance of the swarm to provide a set of baseline results.

  4. Formulate a swarm controller for enhancing group response towards enhancing/facilitating user motion in the smart environment.

Expected Experiences for the Student

• Experience developing experimental test-bed for robot swarm simulation.

• Learning state-of-the-art motion control through time-invariant dynamical systems for mobile robots. 

• Experience virtual robot motion planning for a robotic swarm.

• Experience virtual robot dynamic control.

References:

  1. [1]  L. Huber, A. Billard, and J.-J. Slotine, “Avoidance of Convex and Concave Obstacles With Convergence Ensured Through Contraction,” IEEE Robotics and Automation Letters, vol. 4, no. 2, pp. 1462–1469, 2019. DOI: 10.1109/LRA.2019.2893676

  2. [2]  D. F. Paez-Granados, H. Kadone, and K. Suzuki, “Unpowered Lower-Body Exoskeleton with Torso Lifting Mechanism for Supporting Sit-to-Stand Transitions,” in IEEE International Conference on Intelligent Robots and Systems, (Madrid), pp. 2755–2761, IEEE Xplorer, 2018. DOI:10.1109/IROS.2018.8594199

ProjectMaster Thesis (LASA/BioRob)
PeriodSept.21st, 2021 – Jan. 31st, 2021
Section(s)ME MT SV EL IN MA
Type50% theory, 50% software
Knowledge(s)Programming skills (Python / C++ / Unity)
Control Systems; ROS (a plus); Rviz (a plus)
Subjects(s)Robot control; Control Theory; Simulation; Obstacle Avoidance
Responsible(s)Dr. Diego Paez, Lukas Huber, Dr. Anastasia Bolotnikova (BioRob)
URLCIS project, GitHub code
AvailabilityAvailable