Search Results for author: Marco Hutter

Found 32 papers, 11 papers with code

Data-Efficient Task Generalization via Probabilistic Model-based Meta Reinforcement Learning

no code implementations13 Nov 2023 Arjun Bhardwaj, Jonas Rothfuss, Bhavya Sukhija, Yarden As, Marco Hutter, Stelian Coros, Andreas Krause

We introduce PACOH-RL, a novel model-based Meta-Reinforcement Learning (Meta-RL) algorithm designed to efficiently adapt control policies to changing dynamics.

Meta-Learning Meta Reinforcement Learning +1

Resilient Legged Local Navigation: Learning to Traverse with Compromised Perception End-to-End

no code implementations5 Oct 2023 Jin Jin, Chong Zhang, Jonas Frey, Nikita Rudin, Matias Mattamala, Cesar Cadena, Marco Hutter

In this paper, we model perception failures as invisible obstacles and pits, and train a reinforcement learning (RL) based local navigation policy to guide our legged robot.

Anomaly Detection Navigate +1

MEM: Multi-Modal Elevation Mapping for Robotics and Learning

1 code implementation28 Sep 2023 Gian Erni, Jonas Frey, Takahiro Miki, Matias Mattamala, Marco Hutter

Elevation maps are commonly used to represent the environment of mobile robots and are instrumental for locomotion and navigation tasks.

Colorization Human Detection +1

Learning Risk-Aware Quadrupedal Locomotion using Distributional Reinforcement Learning

no code implementations25 Sep 2023 Lukas Schneider, Jonas Frey, Takahiro Miki, Marco Hutter

Instead of relying on a value expectation, we estimate the complete value distribution to account for uncertainty in the robot's interaction with the environment.

Distributional Reinforcement Learning reinforcement-learning

RAYEN: Imposition of Hard Convex Constraints on Neural Networks

1 code implementation17 Jul 2023 Jesus Tordesillas, Jonathan P. How, Marco Hutter

This paper presents RAYEN, a framework to impose hard convex constraints on the output or latent variable of a neural network.

Fast Traversability Estimation for Wild Visual Navigation

no code implementations15 May 2023 Jonas Frey, Matias Mattamala, Nived Chebrolu, Cesar Cadena, Maurice Fallon, Marco Hutter

We demonstrate the advantages of our approach with experiments and ablation studies in challenging environments in forests, parks, and grasslands.

Navigate Self-Supervised Learning +1

LiDAR-guided object search and detection in Subterranean Environments

no code implementations26 Oct 2022 Manthan Patel, Gabriel Waibel, Shehryar Khattak, Marco Hutter

Detecting objects of interest, such as human survivors, safety equipment, and structure access points, is critical to any search-and-rescue operation.

Disaster Response object-detection +1

Neural Scene Representation for Locomotion on Structured Terrain

no code implementations16 Jun 2022 David Hoeller, Nikita Rudin, Christopher Choy, Animashree Anandkumar, Marco Hutter

We propose a learning-based method to reconstruct the local terrain for locomotion with a mobile robot traversing urban environments.

3D Reconstruction

Advanced Skills through Multiple Adversarial Motion Priors in Reinforcement Learning

no code implementations23 Mar 2022 Eric Vollenweider, Marko Bjelonic, Victor Klemm, Nikita Rudin, Joonho Lee, Marco Hutter

Imitation learning approaches such as adversarial motion priors aim to reduce this problem by encouraging a pre-defined motion style.

Imitation Learning Navigate +2

Learning-based Localizability Estimation for Robust LiDAR Localization

1 code implementation11 Mar 2022 Julian Nubert, Etienne Walther, Shehryar Khattak, Marco Hutter

LiDAR-based localization and mapping is one of the core components in many modern robotic systems due to the direct integration of range and geometry, allowing for precise motion estimation and generation of high quality maps in real-time.

Motion Estimation

Deep Measurement Updates for Bayes Filters

no code implementations1 Dec 2021 Johannes Pankert, Maria Vittoria Minniti, Lorenz Wellhausen, Marco Hutter

In this work, we propose the novel approach Deep Measurement Update (DMU) as a general update rule for a wide range of systems.

Pose Estimation

Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning

3 code implementations24 Sep 2021 Nikita Rudin, David Hoeller, Philipp Reist, Marco Hutter

In this work, we present and study a training set-up that achieves fast policy generation for real-world robotic tasks by using massive parallelism on a single workstation GPU.

reinforcement-learning Reinforcement Learning (RL)

Reconstructing occluded Elevation Information in Terrain Maps with Self-supervised Learning

1 code implementation15 Sep 2021 Maximilian Stölzle, Takahiro Miki, Levin Gerdes, Martin Azkarate, Marco Hutter

We first evaluate a supervised learning approach on synthetic data for which we have the full ground-truth available and subsequently move to several real-world datasets.

Motion Planning Patch Matching +1

3D Surfel Map-Aided Visual Relocalization with Learned Descriptors

no code implementations8 Apr 2021 Haoyang Ye, Huaiyang Huang, Marco Hutter, Timothy Sandy, Ming Liu

In this paper, we introduce a method for visual relocalization using the geometric information from a 3D surfel map.

Camera Relocalization

Imitation Learning from MPC for Quadrupedal Multi-Gait Control

no code implementations26 Mar 2021 Alexander Reske, Jan Carius, Yuntao Ma, Farbod Farshidian, Marco Hutter

We present a learning algorithm for training a single policy that imitates multiple gaits of a walking robot.

Imitation Learning

Articulated Object Interaction in Unknown Scenes with Whole-Body Mobile Manipulation

1 code implementation18 Mar 2021 Mayank Mittal, David Hoeller, Farbod Farshidian, Marco Hutter, Animesh Garg

A kitchen assistant needs to operate human-scale objects, such as cabinets and ovens, in unmapped environments with dynamic obstacles.

Learning a State Representation and Navigation in Cluttered and Dynamic Environments

no code implementations7 Mar 2021 David Hoeller, Lorenz Wellhausen, Farbod Farshidian, Marco Hutter

We show that decoupling the pipeline into these components results in a sample efficient policy learning stage that can be fully trained in simulation in just a dozen minutes.

Representation Learning Visual Navigation

Self-supervised Learning of LiDAR Odometry for Robotic Applications

1 code implementation10 Nov 2020 Julian Nubert, Shehryar Khattak, Marco Hutter

Reliable robot pose estimation is a key building block of many robot autonomy pipelines, with LiDAR localization being an active research domain.


Learning Quadrupedal Locomotion over Challenging Terrain

1 code implementation21 Oct 2020 Joonho Lee, Jemin Hwangbo, Lorenz Wellhausen, Vladlen Koltun, Marco Hutter

The trained controller has taken two generations of quadrupedal ANYmal robots to a variety of natural environments that are beyond the reach of prior published work in legged locomotion.

A Fully-Integrated Sensing and Control System for High-Accuracy Mobile Robotic Building Construction

no code implementations4 Dec 2019 Abel Gawel, Hermann Blum, Johannes Pankert, Koen Krämer, Luca Bartolomei, Selen Ercan, Farbod Farshidian, Margarita Chli, Fabio Gramazio, Roland Siegwart, Marco Hutter, Timothy Sandy

We present a fully-integrated sensing and control system which enables mobile manipulator robots to execute building tasks with millimeter-scale accuracy on building construction sites.

Trajectory Planning

Deep Value Model Predictive Control

no code implementations8 Oct 2019 Farbod Farshidian, David Hoeller, Marco Hutter

The DMPC actor is a Model Predictive Control (MPC) optimizer with an objective function defined in terms of a value function estimated by the critic.

MPC-Net: A First Principles Guided Policy Search

1 code implementation11 Sep 2019 Jan Carius, Farbod Farshidian, Marco Hutter

Our loss function, however, corresponds to the minimization of the control Hamiltonian, which derives from the principle of optimality.

Imitation Learning

Learning agile and dynamic motor skills for legged robots

2 code implementations24 Jan 2019 Jemin Hwangbo, Joonho Lee, Alexey Dosovitskiy, Dario Bellicoso, Vassilios Tsounis, Vladlen Koltun, Marco Hutter

In the present work, we introduce a method for training a neural network policy in simulation and transferring it to a state-of-the-art legged system, thereby leveraging fast, automated, and cost-effective data generation schemes.

reinforcement-learning Reinforcement Learning (RL)

Robust Recovery Controller for a Quadrupedal Robot using Deep Reinforcement Learning

no code implementations22 Jan 2019 Joonho Lee, Jemin Hwangbo, Marco Hutter

We experimentally validate our approach on the quadrupedal robot ANYmal, which is a dog-sized quadrupedal system with 12 degrees of freedom.

Navigate reinforcement-learning +1

Whole-Body Nonlinear Model Predictive Control Through Contacts for Quadrupeds

no code implementations7 Dec 2017 Michael Neunert, Markus Stäuble, Markus Giftthaler, Carmine D. Bellicoso, Jan Carius, Christian Gehring, Marco Hutter, Jonas Buchli

In this work we present a whole-body Nonlinear Model Predictive Control approach for Rigid Body Systems subject to contacts.


Control of a Quadrotor with Reinforcement Learning

1 code implementation17 Jul 2017 Jemin Hwangbo, Inkyu Sa, Roland Siegwart, Marco Hutter

In this paper, we present a method to control a quadrotor with a neural network trained using reinforcement learning techniques.


Robust Visual Place Recognition With Graph Kernels

no code implementations CVPR 2016 Elena Stumm, Christopher Mei, Simon Lacroix, Juan Nieto, Marco Hutter, Roland Siegwart

A novel method for visual place recognition is introduced and evaluated, demonstrating robustness to perceptual aliasing and observation noise.

Visual Place Recognition

Cannot find the paper you are looking for? You can Submit a new open access paper.