toto ANR Entracte Main/Videos


A synergy-based control solution for overactuated characters

Invalid BibTex Entry!

In the current paper, we present a control method based on muscle synergy extraction and adaptation to drive a human arm in a direct dynamics simulation of an overhead throwing motion. The experimental protocol for synergy extraction and model are first presented, followed by a control method consisting of a series of optimizations to adapt muscle parameters and synergies to match experimental data. Results show that the motion can be accurately reproduced thanks to the muscle synergy extraction and adaptation to the model.

Fast contact planning

Invalid BibTex Entry!

Multiped locomotion in cluttered environments is addressed as the problem of planning acyclic sequences of contacts, that characterize the motion. In order to overcome the inherent combinatorial difficulty of the problem, we separate it in two subproblems: first, planning a guide trajectory for the root of the robot and then, generating relevant contacts along this trajectory. This paper proposes theoretical contributions to these two subproblems. We propose a theoretical characterization of the guide trajectory, named ``true feasibility'', which guarantee that a guide can be mapped into the contact manifold of the robot. As opposed to previous approaches, this property makes it possible to assert the relevance of a guide trajectory without explicitly computing contact configurations. This property can be efficiently checked by a sample-based planner (e.g. we implemented a visibility PRM). Since the guide trajectory that we characterized are easily mapped to a valid sequence of contacts, we then focused on how to select a particular sequence with desirable properties, such as robustness, efficiency and naturalness, only considered in cyclic locomotion so far.

Based on these novel theoretical developments, we implemented a complete acyclic contact planner and demonstrate its efficiency by producing a large variety of movements with three very different robots (humanoid, insectoid, dexterous hand) in five challenging scenarios. The planner is very efficient in quality of the produced movements and in computation time: given a computed RB-PRM, a legged figure or a dexterous hand can generate its motion in real time. This result outperforms any previous acyclic contact planner.

Optimization as Motion Selection Principle in Robot Action

Optimization As Motion Selection Principle in Robot Action. Jean-Paul Laumond, Nicolas Mansard and Jean Bernard Lasserre. Commun. ACM, 58(5):64-74, April 2015. (BIBTEX)

Video by Tom Geller (CACM).

Movement is a fundamental characteristic of living systems. Plants and animals must move to survive. Animals are distinguished from plants in that they have to explore the world to feed. The carnivorous plant remains at a fixed position to catch the imprudent insect. Plants must make use of self-centered motions. At the same time the cheetah goes out looking for food. Feeding is a paragon of action. Any action in the physical world requires self-centered movements, exploration movements, or a combination of both. By analogy, a manipulator robot makes use of self-centered motions, a mobile robot moves to explore the world, and a humanoid robot combines both types of motions.

Task efficient contact configurations

Task efficient contact configurations for arbitrary virtual creatures. Steve Tonneau, Julien Pettré and Franck Multon. In Graphics Interface Conference, pages 9-16. Canadian Information Processing Society, 2014. (BIBTEX)

A common issue in three-dimensional animation is the creation of contacts between a virtual creature and the environment. Contacts allow force exertion, which produces motion. This paper addresses the problem of computing contact configurations allowing to perform motion tasks such as getting up from a sofa, pushing an object or climbing. We propose a two-step method to generate contact configurations suitable for such tasks. The first step is an offline sampling of the reachable workspace of a virtual creature. The second step is a run time request confronting the samples with the current environment. The best contact configurations are then selected according to a heuristic for task efficiency. The heuristic is inspired by the force transmission ratio. Given a contact configuration, it measures the potential force that can be exerted in a given direction. Our method is automatic and does not require examples or motion capture data. It is suitable for real time applications and applies to arbitrary creatures in arbitrary environments. Various scenarios (such as climbing, crawling, getting up, pushing or pulling objects) are used to demonstrate that our method enhances motion autonomy and interactivity in constrained environments.

A simple path optimization method for motion planning

A simple path optimization method for motion planning. M. Campana, F. Lamiraux and J-P. Laumond. In IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, September 2015. ([submitted]). (BIBTEX)

Most algorithms in probabilistic sampling-based path planning compute collision-free paths made of straight line segments lying in the configuration space. Due to the randomness of sampling, the paths make detours that need to be optimized. The contribution of this paper is to propose a gradient-based algorithm that transforms a polygonal collision-free path into a shorter one, while both (i) requiring only collision checking, and not any time-consuming obstacle distance computation, and (ii) constraining only part of the configuration variables that may cause a collision, and not the entire configurations. Moreover, the algorithm is simple and requires few parameter tuning. Experimental results include navigation and manipulation tasks, e.g. a PR2 robot working in a kitchen environment and a sliding HRP2 robot.

Whole-body Model-Predictive Control applied to the HRP-2 Humanoid

Experiments with MuJoCo on HRP-2. Jonas Koenemann, Andrea {Del Prete}, Yuval Tassa, Emanuel Todorov, Olivier Stasse, Maren Bennewitz and Nicolas Mansard. In IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, feb 2015. (BIBTEX)

Controlling the robot with a permanently-updated optimal trajectory, also known as model predictive control, is the Holy Grail of whole-body motion generation. Before obtaining it, several challenges should be faced: computation cost, non-linear local minima, algorithm stability, etc. In this paper, we address the problem of applying the updated optimal control in real time on the physical robot. In particular, we focus on the problems raised by the delays due to computation and by the differences between the real robot and the simulated model. Based on the optimal-control solver MuJoCo, we implemented a complete model-predictive controller and we applied it in real time on the physical HRP-2 robot. It is the first time that such a whole-body model predictive controller is applied in real time on a complex dynamic robot. Aside from the technical contributions cited above, the main contribution of this paper is to report the experimental results of this premiere implementation.

Prioritized Optimal Control

Prioritized Optimal Control : a Hierarchical Differential Dynamic Programming approach. Francesco Romano, Andrea {Del Prete}, Nicolas Mansard and Francesco Nori. In IEEE Conference on Robotics and Automation (ICRA), Seattle, USA, may 2015. (BIBTEX)

This paper deals with the generation of motion for complex dynamical systems (such as humanoid robots) to achieve several concurrent objectives. Hierarchy of tasks and optimal control are two frameworks commonly used to this aim. The first one specifies control objectives as a number of quadratic functions to be minimized under strict priorities. The second one minimizes an arbitrary user-defined function of the future state of the system, thus considering its evolution in time. Our recent work on prioritized optimal control merges the advantages of both these methods. This paper reformulates the original prioritized optimal control algorithm with the precise goal of improving its computational speed. We extend the dynamic programming method to work with a hierarchy of tasks. We compared our approach in simulation with both our previous algorithm and classical optimal control. The measured computational improvement represents another step towards the application of prioritized optimal control for online model predictive control of humanoid robots. We believe that this could be the key to unlock the (so far unexploited) dynamic capabilities of these mechanical systems.

A two-stage suboptimal approximation for variable compliance and torque control

A two-stage suboptimal approximation for variable compliance and torque control. P Geoffroy, O Bordron, N Mansard, M Raison, O Stasse and T Bretl. In IEEE European Conference on Control (ECC), Strasbourg, France, June 2014. (BIBTEX)

Variable-stiffness actuator is a very appealing mechatronic design that combines the efficiency of stiff actuator in free space with the consistency of elastic actuation in contact. The control of such an actuation system remains a challenge due to its non-linearity and by the fact that it doubles the number of control inputs. In this paper, we propose an original control strategy to compute the whole-body movement of a complex variable-stiffness robot during dynamic task execution. Operational space control is first used to compute both the joint torque and stiffness from operational references. A non- linear model-predictive controller is then proposed to track at higher frequency these references on each joint separately. The effectiveness of this approach is then validated on two models of real actuator with adjustable stiffness, and finally on an explosive motion to make a humanoid robot jump.

Partial Force Control of Constrained Floating-Base Robots

Partial Force Control of Constrained Floating-Base Robots. Andrea {Del Prete}, Nicolas Mansard, Francesco Nori, Giorgio Metta and Lorenzo Natale. In IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, Chicago, USA, sep 2014. (BIBTEX)

Legged robots are typically in rigid contact with the environment at multiple locations, which add a degree of complexity to their control. We present a method to control the motion and a subset of the contact forces of a floating-base robot. We derive a new formulation of the lexicographic optimization problem typically arising in multi-task motion/force control frameworks. The structure of the constraints of the problem (i.e. the dynamics of the robot) allows us to find a sparse analytical solution. This leads to an equivalent opti- mization with reduced computational complexity, comparable to inverse-dynamics based approaches. At the same time, our method preserves the flexibility of optimization based control frameworks. Simulations were carried out to achieve different multi-contact behaviors on a 23-DoF humanoid robot, validating the presented approach. A comparison with another state-of-the-art control technique with similar computational complexity shows the benefits of our controller, which can eliminate force/torque discontinuities.

Hierarchical quadratic programming

Hierarchical quadratic programming: fast online humanoid-robot motion generation. A Escande, N Mansard and P-B. Wieber. International Journal of Robotic Research (IJRR), 33(7):1006-1028, June 2014. (BIBTEX)

Hierarchical least-square optimization is often used in robotics to inverse a direct function when multiple incompatible objectives are involved. Typical examples are inverse kinematics or dynamics. The objectives can be given as equalities to be satisfied (e.g. point-to-point task) or as areas of satisfaction (e.g. the joint range). This paper proposes a complete solution to solve multiple least-square quadratic problems of both equality and inequality constraints ordered into a strict hierarchy. Our method is able to solve a hierarchy of only equalities ten times faster than the iterative projection hierarchical solvers and can consider inequalities at any level while running at the typical control frequency on whole-body size problems. This generic solver is used to resolve the redundancy of humanoid robots while generating complex movements in constrained environment.

A Simple Method to Calibrate Kinematical Invariants: Application to Overhead Throwing

A simple method to calibrate kinematical invariants: application to overhead throwing. A. Muller, C. Germain, C. Pontonnier and G. Dumont. In International Conference on Biomechanics in Sports (ISBS), Poitiers, France, June 2015. (BIBTEX)

The aim of this paper is to present a simple calibration method aimed at optimizing the kinematical invariants of a whole body motion capture model, meaning limb lengths and some of the marker placements. A case study and preliminary results are presented and give encouraging insights about the generalized use of such a method in motion analysis in sports.

Motion capture of throwing movements (7m): preliminary details about the experiment

Motion capture of Parkour athletes: preliminary details about the experiment