.gitignore

+.vscode
\ No newline at end of file

.gitlab-ci.yml

 image: memmos.laas.fr:5000/gepetto/articles
 
-pages:
+build:
+  stage: build
   script:
-    - make
+    - make all
   artifacts:
     paths:
       - public
+
+test:
+  stage: test
+  script:
+    - make check
+
+deploy:
+  stage: deploy
+  only:
+    refs:
+      - main@gepetto/articles
+  before_script:
+    - eval $(ssh-agent -s)
+    - echo "$SSH_PRIVATE_KEY" | ssh-add -
+  script:
+    - make deploy

Dockerfile

-FROM ubuntu:20.04
+FROM pandoc/core:latest-ubuntu
 
-RUN --mount=type=cache,sharing=locked,target=/var/cache/apt --mount=type=cache,sharing=locked,target=/var/lib/apt \
-    apt-get update -qqy && DEBIAN_FRONTEND=noninteractive apt-get install -qqy \
-    pandoc \
-    pandoc-citeproc \
-    make
+ENTRYPOINT []
+
+
+RUN --mount=type=cache,sharing=locked,target=/var/cache/apt \
+    --mount=type=cache,sharing=locked,target=/var/lib/apt \
+    apt-get update -y && DEBIAN_FRONTEND=noninteractive apt-get install -qqy \
+    make \
+    openssh-client \
+    python3-requests \
+    rsync
+
+RUN addgroup --gid 1110 gepetto \
+ && adduser --ingroup gepetto --disabled-password --uid 5495 gsaurel
+USER gsaurel
+RUN mkdir -p ~/.ssh \
+ && ssh-keyscan "memmos.laas.fr" > ~/.ssh/known_hosts \
+ && chmod 700 ~/.ssh \
+ && chmod 644 ~/.ssh/known_hosts

LICENSE

+BSD 2-Clause License
+
+Copyright (c) 2022, CNRS
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+   list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice,
+   this list of conditions and the following disclaimer in the documentation
+   and/or other materials provided with the distribution.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

Makefile

@@ -4,14 +4,20 @@ DEST="/net/cubitus/projects/Partage_GEPETTO/Public/articles"
 
 all: ${OUTPUTS}
 
-public/%.html: %.md template.html bibliography.bib
-	pandoc -s -F pandoc-citeproc --template template.html -o $@ $<
+public/%.html: %.md template.html
+	pandoc -s \
+		-F ./bibtex.py \
+		--template template.html \
+		--fail-if-warnings \
+		-o $@ $<
+
+check: all
+	! grep '/videos/watch' ${SOURCES}  # you have to replace "watch" by "embed"
+	! grep '/w/' ${SOURCES}  # you need an embed URL link
+
+deploy: check
+	chmod a+r,g+w ${OUTPUTS}
+	rsync -avzP --delete public/ gepetto-deploy@memmos.laas.fr:${DEST}
 
 clean:
 	rm -f ${OUTPUTS}
-
-deploy: all
-	mkdir -p ${DEST}
-	cp ${OUTPUTS} ${DEST}
-	chown -R :gepetto ${DEST}
-	chmod -R a+rX,g+w ${DEST}

README.md

 # Single page article presentation
+
+## Create your own
+
+You can use `cosyslam.md` as a template. Just copy it under a different `*.md` name, and replace what you want inside.
+
+## Publish
+
+Publication https://gepettoweb.laas.fr/articles/ is done via gitlab CI/CD on the main branch
+
+## Build existing pages
+
+This is not mandatory, as gitlab CI/CD is doing all the required chores, but if you want to build things locallay:
+
+- with pandoc >= 2.14: `make`
+- otherwise: `docker run --rm -v $PWD:/data -it memmos.laas.fr:5000/gepetto/articles make`

amit_icra_22.md

+---
+title: "Value learning from trajectory optimization and Sobolev descent: A step toward reinforcement learning with superlinear convergence properties"
+subtitle: IEEE ICRA - International Conference on Robotics and Automation, 2022
+author:
+- Amit Parag ^1,2^
+- Sébastien Kleff ^1,3^
+- Léo Saci ^1^
+- <a href="https://gepettoweb.laas.fr/index.php/Members/NicolasMansard">Nicolas Mansard</a> ^1,2^
+- <a href="https://homepages.laas.fr/ostasse/drupal/">Olivier Stasse</a> ^1,2^
+org:
+- ^1^ Gepetto Team, LAAS-CNRS, France.
+- ^2^ Artificial and Natural Intelligence Toulouse Institute, Toulouse.
+- ^3^ New York University, USA
+
+hal: hal-03356261
+peertube: https://peertube.laas.fr/videos/embed/00e812a0-7321-4f05-8ab8-3250f2a49deb
+code: https://github.com/amitparag/Kuka-arm-DvP
+...
+
+## Abstract
+
+The recent successes in deep reinforcement learning largely rely on the capabilities of generating masses of data, which in turn implies the use of a simulator.
+In particular, current progress in multi body dynamic simulators are underpinning the implementation of reinforcement learning for end-to-end control of robotic systems.
+Yet simulators are mostly considered as black boxes while we have the knowledge to make them produce a richer information.
+In this paper, we are proposing to use the derivatives of the simulator to help with the convergence of the learning.
+For that, we combine model-based trajectory optimization to produce informative trials using 1st- and 2nd-order simulation derivatives.
+These locally-optimal runs give fair estimates of the value function and its derivatives, that we use to accelerate the convergence of the critics using Sobolev learning.
+We empirically demonstrate that the algorithm leads to a faster and more accurate estimation of the value function.
+The resulting value estimate is used in model-predictive controller as a proxy for shortening the preview horizon.
+We believe that it is also a first step toward superlinear reinforcement learning algorithm using simulation derivatives, that we need for end-to-end legged locomotion.

assirelli_2022.md

+---
+title: "Whole-Body MPC Without Foot References for the Locomotion of an Impedance Controlled Robot"
+subtitle: Presented at 2022 IEEE ICRA - International Conference on Robotics and Automation
+author:
+- Alessandro Assirelli^1^
+- Fanny Risbourg ^1^
+- Gianni Lunardi ^2^
+- Thomas Flayols ^1,3^
+- Nicolas Mansard ^1,3^
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse, France
+- ^2^ Industrial Engineering Department, University of Trento, Italy
+- ^3^ Artificial and Natural Intelligence Toulouse Institute, Toulouse, France
+hal: hal-03778738
+peertube: https://peertube.laas.fr/videos/embed/f8780cf3-0a9a-4b76-adac-400f822c1b26
+...
+
+## Abstract
+With the fast progress of quadruped robots, we also see the rise of advanced controllers able to take whole-body decisions without any model reduction. Recently, whole-body model predictive control have been demonstrated on several legged robots. Based on these results, this paper presents a novel walking controller. Contrary to previously demonstrated approaches, our controller does not require the pre-computation of guide trajectories for the foot, nor of specific foot  location, but rather decides on the flight the best foot movement using an original cost formulation. The predictive controller is then applied at the actuator level using an impedance controller, without requiring the more costly low-level torque controller that previous methods used. The method is validated on the real robot Solo, using an open-source implementation based on the solver Crocoddyl. We evaluate in depth the quality of the produced walk, despite external disturbance, and provide longer experiments in the companion video.

batto2023.md

+---
+title: "Comparative metrics of advanced serial/parallel biped design and characterization of the main contemporary architectures"
+subtitle: Presented at IEEE-RAS International Conference on Humanoid Robots 2023
+author:
+- Virgile BATTO ^1,2^
+- Thomas Flayols ^1,3^
+- Nicolas Mansard ^1,3^
+- Margot Vulliez ^2^
+
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse, Toulouse, France
+- ^2^ Auctus, Inria, centre de l’Université de Bordeaux, Talence, France
+- ^3^ Artificial and Natural Intelligence Toulouse Institute, Toulouse, France
+hal: hal-04191553
+peertube: https://peertube.laas.fr/videos/embed/4690941b-eb73-45fe-bdc6-381136d67c8a
+...
+
+## Abstract
+
+The best achievements in bipedal locomotion have resulted from associating an intelligent and efficient design with clever and robust control.
+While several control frameworks exist, the design of the legs of our biped robots still lacks a systematic approach and remains a crucial challenge for robot mobility.
+This paper introduces several criteria to characterize the design of bipedal legs.
+They aim to guide the design choices and could be implemented in a codesign approach.
+They reflect the leg overall performances (ability to produce dynamic and accurate foot movements, absorb impacts, lower the motor torques needed to stand up) and characterize the design compactness.
+We give the algorithmic formulations to evaluate them beyond classical serial designs, to account for any parallel mechanisms.
+To validate these criteria, we developed a library of open-source CAD models describing the main existing biped architectures, which can be used as a database for future design studies.
+We discuss the comparative performances of these architectures.
+We hope this quantified discussion can serve as a baseline to better design future biped robots.

bibliography.bib

-// TODO: update this with HAL entry
-@InProceedings{CosySLAM,
-  author={César Debeunne and Médéric Fourmy and Yann Labbé and Pierre-Alexandre Léziart and Guilhem Saurel and Joan
-    Solà and Nicolas Mansard},
-  title={CosySLAM: tracking contact features using visual-inertial object-level SLAM for locomotion},
-  booktitle={ICRA (submitted)},
-  year={2022}
-}

bibtex.py

+#!/usr/bin/env python3
+"""
+Pandoc filter to include a raw bibtex entry from HAL
+"""
+
+import json
+import sys
+
+import requests
+
+API = "https://api.archives-ouvertes.fr/search/"
+
+
+def bibtex():
+    doc = json.load(sys.stdin)
+    if "hal" in doc["meta"]:
+        hal = doc["meta"]["hal"]["c"][0]["c"]
+        bib = requests.get(API, {"q": f"halId_id:{hal}", "wt": "bibtex"})
+        bib.raise_for_status()
+        doc["meta"]["bibtex"] = {
+            "t": "MetaBlocks",
+            "c": [
+                {"t": "CodeBlock", "c": [["", ["bibtex"], []], bib.content.decode()]}
+            ],
+        }
+        url = requests.get(API, {"q": f"halId_id:{hal}"})
+        url.raise_for_status()
+        doc["meta"]["hal_url"] = {
+            "t": "MetaString",
+            "c": url.json()["response"]["docs"][0]["uri_s"],
+        }
+    json.dump(doc, sys.stdout)
+
+
+if __name__ == "__main__":
+    bibtex()

bonnefoy_otmani_2023.md

+
+---
+title: Modelisation of a Human-Exoskeleton Interaction for Cerebral Palsy
+subtitle: 2023 International Symposium on Medical Robotics (ISMR)
+author:
+- Aurélie BONNEFOY ^1^
+- Sabrina OTMANI ^1^,^2^
+- <a href="https://gepettoweb.laas.fr/index.php/Members/NicolasMansard">Nicolas MANSARD</a> ^1^
+- <a href="https://homepages.laas.fr/ostasse/drupal/">Olivier STASSE</a> ^1^
+- <a href="https://ica.cnrs.fr/author/gmichon/">Guilhem MICHON</a> ^2^
+- <a href="https://gepettoweb.laas.fr/index.php/Members/BrunoWatier">Bruno WATIER</a> ^1^
+org:
+- ^1^ Gepetto Team, LAAS-CNRS, France.
+- ^2^ MS2M Team, ICA, Toulouse.
+
+code: https://github.com/ABonnefoy/Exo_CP
+peertube: https://peertube.laas.fr/videos/embed/af8e5014-ee45-43e6-8db7-8bcd26dc648b?autoplay=1&amp;title=0&amp;warningTitle=0&amp;p2p=0
+...
+
+## Abstract
+
+This paper presents a method to model a human-
+exoskeleton interaction for patients suffering from cerebral
+palsy (CP). More precisely a model of the gait related to
+spastic CP is proposed using an optimization program based
+on experimental data. The model is done using mechanical
+differential equations of motion. A unique feature of this paper
+is the Clinical Gait Analysis (CGA) performed on two 9 years
+old twin sisters. One has spastic cerebral palsy (C) while the
+other is healthy (H) thus without any impairment. This paper
+aims at determining the proportion of the walking efforts that
+can be supported by the exoskeleton in order to allow a CP
+child gait to converge toward a non-pathological one. For this
+purpose, minimal torques produced by the human in interaction
+with the exoskeleton were studied. The interaction between
+the human and the exoskeleton is realized using optimisation
+methods such as SLSQP and QuadProg. Ground contacts are
+also included in the modelisation. Results show that the human
+produces joint torques in the same range of the ones of C.
+Exoskeleton succeds in producing additionnal torques to lead
+the pathological gait to a non-pathological one. The code for
+running the simulations is available on git.

cat.md

+---
+title: "CaT: Constraints as Terminations for Legged Locomotion Reinforcement Learning"
+subtitle: Check <https://constraints-as-terminations.github.io> for code, article and videos!
+author:
+- Elliot Chane-Sane ^1^
+- Pierre-Alexandre Leziart ^1^
+- Thomas Flayols ^1^
+- Olivier Stasse ^1,2^
+- Philippe Souères ^1^
+- Nicolas Mansard ^1,2^
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse
+- ^2^ Artificial and Natural Intelligence Toulouse Institute, Toulouse
+hal: hal-04523167
+code: https://github.com/gepetto/constraints-as-terminations
+...
+
+## Abstract
+
+Deep Reinforcement Learning (RL) has demonstrated impressive results in solving complex robotic tasks such as quadruped locomotion. Yet, current solvers fail to produce efficient policies respecting hard constraints. In this work, we advocate for integrating constraints into robot learning and present Constraints as Terminations (CaT), a novel constrained RL algorithm. Departing from classical constrained RL formulations, we reformulate constraints through stochastic terminations during policy learning: any violation of a constraint triggers a probability of terminating potential future rewards the RL agent could attain. We propose an algorithmic approach to this formulation, by minimally modifying widely used off-the-shelf RL algorithms in robot learning (such as Proximal Policy Optimization). Our approach leads to excellent constraint adherence without introducing undue complexity and computational overhead, thus mitigating barriers to broader adoption. Through empirical evaluation on the real quadruped robot Solo crossing challenging obstacles, we demonstrate that CaT provides a compelling solution for incorporating constraints into RL frameworks.
+ 
+Videos and code are available at <https://constraints-as-terminations.github.io>
+

cosyslam.md

 ---
-title: "CosySLAM: tracking contact features using visual-inertial object-level SLAM for locomotion"
-subtitle: Submitted to 2022 IEEE ICRA - International Conference on Robotics and Automation
+title: "CosySlam: investigating object-level SLAM for detecting locomotion surfaces"
+subtitle: Submitted to 2022 IEEE/RSJ IROS - International Conference on Intelligent Robots and Systems
 author:
-- César Debeunne ^1^
-- Médéric Fourmy ^1,2^
+- César Debeunne ^1,2^
+- Médéric Fourmy ^2^
 - Yann Labbé ^3^
-- Pierre-Alexandre Léziart ^1^
-- Guilhem Saurel ^1^
-- Joan Solà ^1,3^
-- <a href="https://gepettoweb.laas.fr/index.php/Members/NicolasMansard">Nicolas Mansard</a> ^1,2^
+- Pierre-Alexandre Léziart ^2^
+- Guilhem Saurel ^2^
+- Joan Solà ^2,4^
+- <a href="https://gepettoweb.laas.fr/index.php/Members/NicolasMansard">Nicolas Mansard</a> ^2,5^
 org:
-- ^1^ LAAS-CNRS, Université de Toulouse
-- ^2^ Artificial and Natural Intelligence Toulouse Institute, Toulouse
-- ^3^ Intitut de Robòtica i Informàtica Industrial, Barcelona
-- ^4^ Inria, École normale supérieure, CNRS, PSL Research University, Paris
-cite: CosySLAM
-paper: https://hal.laas.fr/hal-TODO/document
-code: https://github.com/gepetto/TODO
-peertube: https://peertube.laas.fr/videos/embed/a78430ea-09b0-4e31-8ca6-7436c7af9165
+- ^1^ ISAE-Supaero, Toulouse
+- ^2^ LAAS-CNRS, Université de Toulouse
+- ^3^ Inria, École normale supérieure, CNRS, PSL Research University, Paris
+- ^4^ Intitut de Robòtica i Informàtica Industrial, Barcelona
+- ^5^ Artificial and Natural Intelligence Toulouse Institute, Toulouse
+hal: hal-03351438
+peertube: https://peertube.laas.fr/videos/embed/56edb26d-e2c3-46ac-909b-61f55d10c569
+data: https://gepettoweb.laas.fr/data/cosyslam/
 ...
 
 ## Abstract
 
-A legged robot is equipped with several sensors observing different classes of information, in order to provide various estimates on its states and its environment.
-While state estimation and mapping in this domain have traditionally been investigated through multiple local filters, recent progresses have been made toward tightly-coupled estimation.
-Multiple observations are then merged into an a-posteriori maximum estimating several quantities that otherwise were separately estimated.
-With this paper, our goal is to move one step further, by leveraging on object-based simultaneous localization and mapping.
-We use an object pose estimator to localize the relative placement of the robot with respect to large elements of the environments, e.g. stair steps.
-These measurements are merged with other typical observations of legged robots, e.g. inertial measurements, to provide an estimation of the robot state (position, orientation and velocity of the basis) along with an accurate estimation of the environment pieces.
-It then provides a consistent estimation of these two quantities, which is an important property as both would be needed to control the robot locomotion.
-We provide a complete implementation of this idea with the object tracker CosyPose, which we trained on our environment and for which we provide a covariance model, and with the SLAM engine Wolf used as a visual-inertial estimator on the quadruped robot Solo.
+While blindfolded legged locomotion has demonstrated impressive capabilities in the last few years, further progresses
+are expected from using exteroceptive perception to better adapt the robot behavior to the available surfaces of
+contact. In this paper, we investigate whether mono cameras are suitable sensors for that aim. We propose to rely on
+object-level SLAM, fusing RGB images and inertial measurements, to simultaneously estimate the robot balance state
+(orientation in the gravity field and velocity), the robot position, and the location of candidate contact surfaces. We
+used CosyPose, a learning-based object pose estimator for which we propose an empirical uncertainty model, as the sole
+front-end of our visual inertial SLAM. We then combine it with inertial measurements which ideally complete the system
+observability, although extending the proposed approach would be straightforward (e.g. kinematic information about the
+contact, or a feature based visual front end). We demonstrate the interest of object-based SLAM on several locomotion
+sequences, by some absolute metrics and in comparison with other mono SLAM.

dantec22.md

+---
+title: "First Order Approximation of Model Predictive Control Solutions for High Frequency Feedback"
+subtitle: Submitted to IEEE RA-L - Robotics and Automation Letter and IEEE ICRA 2022 - International Conference on Robotics and Automation
+author:
+- Ewen Dantec ^1,2^
+- Michel Taïx ^1^
+- <a href="https://gepettoweb.laas.fr/index.php/Members/NicolasMansard">Nicolas Mansard</a> ^1,2^
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse
+- ^2^ Artificial and Natural Intelligence Toulouse Institute, Toulouse
+hal: hal-03419712
+peertube: https://peertube.laas.fr/videos/embed/b71febbc-d5f4-4e35-a339-b8a3cbd6fed1
+...
+
+## Abstract
+
+The lack of computational power on mobile robots is a well-known challenge when it comes to implementing a real-time MPC scheme to perform complex motions.
+Currently the best solvers are barely able to reach 100Hz for computing the control of a whole-body legged model, while modern robots are expecting new torque references in less than 1ms.
+This problem is usually tackled by using a handcrafted low-level tracking control whose inputs are the low-frequency trajectory computed by the MPC.
+We show that a linear state feedback controller naturally arises from the optimal control formulation and can be used directly in the low-level control loop along with other sensitivities of relevant time-varying parameters of the problem.
+When the optimal control problem is solved by DDP, this linear controller can be computed for cheap as a by-product of the backward pass, and corresponds in part to the classical Riccati gains.
+A side effect of our proposition is to show that Riccati gains are valuable assets that must be used to achieve an efficient control and that they are not stiffer than the optimal control scheme itself.
+We propose a complete implementation of this idea on a full-scale humanoid robot and demonstrate its importance with real experiments on the robot Talos.

dantechumanoid22.md

+---
+title: "Whole-Body Model Predictive Control for Biped Locomotion on a Torque-Controlled Humanoid Robot"
+subtitle: Submitted to IEEE-RAS International Conference on Humanoid Robots 2022
+author:
+- Ewen Dantec ^1,2^
+- Maximilien Naveau ^1^
+- Pierre Fernbach ^3^
+- Nahuel Villa ^1^
+- Guilhem Saurel ^1^
+- Olivier Stasse ^1^
+- Michel Taïx ^1^
+- <a href="https://gepettoweb.laas.fr/index.php/Members/NicolasMansard">Nicolas Mansard</a> ^1,2^
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse
+- ^2^ Artificial and Natural Intelligence Toulouse Institute, Toulouse
+- ^3^ TOWARD, Toulouse, France
+hal: hal-03724019
+peertube: https://peertube.laas.fr/videos/embed/caf150bf-878b-40d9-88db-710b71730f9f
+data: https://gepettoweb.laas.fr/data/wbwalk/
+...
+
+## Abstract
+
+ In this paper, we present a whole-body Model Predictive Control framework for locomotion and validate it on the humanoid robot Talos. Using a time horizon of 1.5 second and a
+20 Degree of Freedom model, the proposed controller outputs the optimal feedforward torque and Riccati-based feedback policy at a frequency of 100 Hz and the optimal feedback
+torque at 2kHz. Contact constraints are handled through wrench regularization following a normal force reference in order to
+hint smooth force transitions to the solver. Contact locations and timings are user-defined, and Bezier curves are implemented
+as reference feet trajectories. Experimental validation includes dynamic locomotion at different gaits as well as 10 cm height
+stairstep crossing. To the best of the authors’ knowledge, this experimental result marks the first achievement of locomotion
+on non-flat terrain for an electric torque-controlled humanoid robot using a full-dynamics Model Predictive Control scheme.

fadini2021.md

+---
+title: "Computational design of energy-efficient legged robots: Optimizing for size and actuators"
+subtitle: IEEE International Conference on Robotics and Automation (ICRA 2021), May 2021, Xian, China.
+author:
+- Gabriele Fadini ^1^
+- Thomas Flayols ^1^
+- Andrea del Prete ^2^
+- Nicolas Mansard ^1,3^
+- Philippe Souères ^1^
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse
+- ^2^ University of Trento, Industrial Engineering Department, Trento, Italy
+- ^3^ Artificial and Natural Intelligence Toulouse Institute, Toulouse
+hal: hal-02993624
+peertube: https://peertube.laas.fr/videos/embed/b22VxXJRJokKuj52aVEU4N
+...
+
+## Abstract
+
+This paper presents a computational framework for the design of high-performance legged robotic systems. The framework relies on the concurrent optimization of hardware parameters and control trajectories to find the best robot design for a given task. In particular, we focus on energy efficiency, presenting novel electro-mechanical models to account for the losses of the actuators due to friction and Joule effects. Thanks to a bi-level optimization scheme, featuring a genetic algorithm in the outer loop, our framework can also optimize for the duration of the motion, the actuators, and the size of the robot. We present a novel approach to scale both the actuators and the robot structure in a way that ensures structural integrity by maintaining constant the normalized deflection of the links. We validated our approach by designing a two-joint monoped robot to execute a jumping task. Our results show that our framework can lead to remarkable energy savings (up to 60%) thanks to the concurrent optimization of robot size, motion duration, and actuators.

fadini2022.md

+---
+title: "Simulation aided co-design for robust robot optimization"
+subtitle: Published in IEEE Robotics and Automation Letters and presented at 2022 IEEE/RSJ IROS - International Conference on Intelligent Robots and Systems
+author:
+- Gabriele Fadini ^1^
+- Thomas Flayols ^1^
+- Andrea del Prete ^2^
+- Philippe Souères ^1^
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse
+- ^2^ University of Trento, Industrial Engineering Department, Trento, Italy
+hal: hal-03592085
+peertube: https://peertube.laas.fr/videos/embed/7z69ih2EmqjQGF157h1JBt
+...
+
+## Abstract
+
+This paper presents a computational framework for the design of high-performance legged robotic systems. The framework relies on the concurrent optimization of hardware parameters and control trajectories to find the best robot design for a given task. In particular, we focus on energy efficiency, presenting novel electro-mechanical models to account for the losses of the actuators due to friction and Joule effects. Thanks to a bi-level optimization scheme, featuring a genetic algorithm in the outer loop, our framework can also optimize for the duration of the motion, the actuators, and the size of the robot. We present a novel approach to scale both the actuators and the robot structure in a way that ensures structural integrity by maintaining constant the normalized deflection of the links. We validated our approach by designing a two-joint monoped robot to execute a jumping task. Our results show that our framework can lead to remarkable energy savings (up to 60%) thanks to the concurrent optimization of robot size, motion duration, and actuators.

fadini2023.md

+---
+title: "Co-designing versatile quadruped robots <br/> for dynamic and energy-efficient motions"
+subtitle: Pre-print version, paper submitted to TRO
+author:
+- Gabriele Fadini ^1^
+- Shivesh Kumar ^2^
+- Rohit Kumar ^2^
+- Thomas Flayols ^1^
+- Andrea del Prete ^3^
+- Justin Carpentier ^4^
+- Philippe Souères ^1^
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse
+- ^2^ Robotics Innovation Center, DFKI, Bremen, Germany
+- ^3^ University of Trento, Industrial Engineering Department, Trento, Italy
+- ^4^ INRIA and Departement d'informatique de l'ENS, Paris, France
+hal: hal-04162737
+peertube: https://peertube.laas.fr/videos/embed/90fe2efb-aa5e-4461-aad8-3df1b0862107
+...
+
+## Abstract
+This paper presents a bi-level optimization frame-work to concurrently optimize a quadruped hardware and control policies for achieving dynamic cyclic behaviors. The long-term vision to drive the design of dynamic and efficient robots by means of computational techniques is applied to improve the development of a new quadruped prototype. The scale of the robot and its actuators are optimized for energy efficiency considering a complete model of the motor, that includes friction, torque, and bandwidth limitations. This model is used to optimize the power consumption during bounding and backflip tasks and is validated by tracking the output trajectories on the first prototype iteration. The co-design results show an improvement of up to 87% for a single task optimization. It appears that, for jumping forward, robots with longer thighs perform better, while for backflips, longer shanks are better suited. To understand the trade-off between these different choices, a Pareto set is constructed to guide the design of the next prototype.

figaroh-2023.md

+---
+title: "FIGAROH: a Python toolbox for dynamic identification and geometric calibration of robots and humans"
+author:
+- Dinh Vinh Thanh Nguyen ^1,2^,
+- Vincent Bonnet ^1,3^,
+- Maxime Sabbah ^1^,
+- Maxime Gautier ^4^,
+- Pierre Fernbach ^2^,
+- Florent Lamiraux ^1^
+
+org:
+- ^1^ LAAS-CNRS, Université de Toulouse, Toulouse, France,
+- ^2^ Toward S.A.S., Toulouse, France,
+- ^3^ IUT Paul Sabatier - Université de Toulouse, Toulouse, France,
+- ^4^ LS2N, Université de Nantes, Nantes, France.
+
+peertube: https://peertube.laas.fr/videos/embed/65291dd2-47a2-4357-bb59-05f1f848acb7
+...
+
+## Abstract
+
+The accuracy of the geometric and dynamic models for robots and humans is crucial for simulation, control, and motion analysis. For example, joint torque, which is a function of geometric and dynamic parameters, is a critical variable that heavily impacts the performance of model-based control, or that can motivate a clinical decision after a biomechanical analysis. Fortunately, these models can be identified using extensiveworks from literature. However, for a non-expert, building an identification model and designing an experimentation plan,which should not require long hours and/or lead to poor results, is not a trivial task, especially for anthropometric structuressuch as humanoids or humans that need frequent update. In this work, we propose a unified framework for geometric calibration and dynamic identification in the form of a Python open-source toolbox. Besides identification model building and data processing, the toolbox can automatically generate exciting postures and motions to minimize the experimental burden from the robot, measurements, and environment description.The possibilities of this toolbox are exemplified with several datasets of human, humanoid, and serial robots.
+