Lionel Ott
YOU?
Author Swipe
View article: Towards Fast and Scalable Normal Integration using Continuous Components
Towards Fast and Scalable Normal Integration using Continuous Components Open
Surface normal integration is a fundamental problem in computer vision, dealing with the objective of reconstructing a surface from its corresponding normal map. Existing approaches require an iterative global optimization to jointly estim…
View article: CueLearner: Bootstrapping and local policy adaptation from relative feedback
CueLearner: Bootstrapping and local policy adaptation from relative feedback Open
Human guidance has emerged as a powerful tool for enhancing reinforcement learning (RL). However, conventional forms of guidance such as demonstrations or binary scalar feedback can be challenging to collect or have low information content…
View article: Learning Affordances from Interactive Exploration using an Object-level Map
Learning Affordances from Interactive Exploration using an Object-level Map Open
Many robotic tasks in real-world environments require physical interactions with an object such as pick up or push. For successful interactions, the robot needs to know the object's affordances, which are defined as the potential actions t…
View article: NeuSurfEmb: A Complete Pipeline for Dense Correspondence-based 6D Object Pose Estimation without CAD Models
NeuSurfEmb: A Complete Pipeline for Dense Correspondence-based 6D Object Pose Estimation without CAD Models Open
State-of-the-art approaches for 6D object pose estimation assume the availability of CAD models and require the user to manually set up physically-based rendering (PBR) pipelines for synthetic training data generation. Both factors limit t…
View article: Task Adaptation in Industrial Human-Robot Interaction: Leveraging Riemannian Motion Policies
Task Adaptation in Industrial Human-Robot Interaction: Leveraging Riemannian Motion Policies Open
In real-world industrial environments, modern robots often rely on human operators for crucial decision-making and mission synthesis from individual tasks. Effective and safe collaboration between humans and robots requires systems that ca…
View article: Waverider: Leveraging Hierarchical, Multi-Resolution Maps for Efficient and Reactive Obstacle Avoidance
Waverider: Leveraging Hierarchical, Multi-Resolution Maps for Efficient and Reactive Obstacle Avoidance Open
Fast and reliable obstacle avoidance is an important task for mobile robots. In this work, we propose an efficient reactive system that provides high-quality obstacle avoidance while running at hundreds of hertz with minimal resource usage…
View article: A framework for collaborative multi-robot mapping using spectral graph wavelets
A framework for collaborative multi-robot mapping using spectral graph wavelets Open
The exploration of large-scale unknown environments can benefit from the deployment of multiple robots for collaborative mapping. Each robot explores a section of the environment and communicates onboard pose estimates and maps to a centra…
View article: Team CERBERUS Wins the DARPA Subterranean Challenge: Technical Overview and Lessons Learned
Team CERBERUS Wins the DARPA Subterranean Challenge: Technical Overview and Lessons Learned Open
This article presents the CERBERUS robotic system-of-systems, which won the DARPA Subterranean Challenge Final Event in 2021. The Subterranean Challenge was organized by DARPA with the vision to facilitate the novel technologies necessary …
View article: Zero123-6D: Zero-shot Novel View Synthesis for RGB Category-level 6D Pose Estimation
Zero123-6D: Zero-shot Novel View Synthesis for RGB Category-level 6D Pose Estimation Open
Estimating the pose of objects through vision is essential to make robotic platforms interact with the environment. Yet, it presents many challenges, often related to the lack of flexibility and generalizability of state-of-the-art solutio…
View article: To Fuse or Not to Fuse: Measuring Consistency in Multi-Sensor Fusion for Aerial Robots
To Fuse or Not to Fuse: Measuring Consistency in Multi-Sensor Fusion for Aerial Robots Open
Aerial vehicles are no longer limited to flying in open space: recent work has focused on aerial manipulation and up-close inspection. Such applications place stringent requirements on state estimation: the robot must combine state informa…
View article: Soliro -- a hybrid dynamic tilt-wing aerial manipulator with minimal actuators
Soliro -- a hybrid dynamic tilt-wing aerial manipulator with minimal actuators Open
The ability to enter in contact with and manipulate physical objects with a flying robot enables many novel applications, such as contact inspection, painting, drilling, and sample collection. Generally, these aerial robots need more degre…
View article: Chasing millimeters: design, navigation and state estimation for precise in-flight marking on ceilings
Chasing millimeters: design, navigation and state estimation for precise in-flight marking on ceilings Open
Precise markings for drilling and assembly are crucial, laborious construction tasks. Aerial robots with suitable end-effectors are capable of markings at the millimeter scale. However, so far, they have only been demonstrated under labora…
View article: Open X-Embodiment: Robotic Learning Datasets and RT-X Models
Open X-Embodiment: Robotic Learning Datasets and RT-X Models Open
Large, high-capacity models trained on diverse datasets have shown remarkable successes on efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led to a consolidation of pretrained models, with gen…
View article: Self-Supervised Learning for Interactive Perception of Surgical Thread for Autonomous Suture Tail-Shortening
Self-Supervised Learning for Interactive Perception of Surgical Thread for Autonomous Suture Tail-Shortening Open
Accurate 3D sensing of suturing thread is a challenging problem in automated surgical suturing because of the high state-space complexity, thinness and deformability of the thread, and possibility of occlusion by the grippers and tissue. I…
View article: Efficient volumetric mapping of multi-scale environments using wavelet-based compression
Efficient volumetric mapping of multi-scale environments using wavelet-based compression Open
Volumetric maps are widely used in robotics due to their desirable properties in applications such as path planning, exploration, and manipulation.Constant advances in mapping technologies are needed to keep up with the improvements in sen…
View article: Efficient volumetric mapping of multi-scale environments using wavelet-based compression
Efficient volumetric mapping of multi-scale environments using wavelet-based compression Open
Volumetric maps are widely used in robotics due to their desirable properties in applications such as path planning, exploration, and manipulation. Constant advances in mapping technologies are needed to keep up with the improvements in se…
View article: Efficient volumetric mapping of multi-scale environments using wavelet-based compression
Efficient volumetric mapping of multi-scale environments using wavelet-based compression Open
Volumetric maps are widely used in robotics due to their desirable properties in applications such as path planning, exploration, and manipulation. Constant advances in mapping technologies are needed to keep up with the improvements in se…
View article: Automatic extension of a symbolic mobile manipulation skill set
Automatic extension of a symbolic mobile manipulation skill set Open
Symbolic planning can provide an intuitive interface for non-expert users to operate autonomous robots by abstracting away much of the low-level programming. However, symbolic planners assume that the initially provided abstract domain and…
View article: Material-agnostic Shaping of Granular Materials with Optimal Transport
Material-agnostic Shaping of Granular Materials with Optimal Transport Open
From construction materials, such as sand or asphalt, to kitchen ingredients, like rice, sugar, or salt; the world is full of granular materials. Despite impressive progress in robotic manipulation, manipulating and interacting with granul…
View article: Neural Implicit Vision-Language Feature Fields
Neural Implicit Vision-Language Feature Fields Open
Recently, groundbreaking results have been presented on open-vocabulary semantic image segmentation. Such methods segment each pixel in an image into arbitrary categories provided at run-time in the form of text prompts, as opposed to a fi…
View article: Robust Sampling-Based Control of Mobile Manipulators for Interaction With Articulated Objects
Robust Sampling-Based Control of Mobile Manipulators for Interaction With Articulated Objects Open
International audience
View article: Obstacle avoidance using Raycasting and Riemannian Motion Policies at kHz rates for MAVs
Obstacle avoidance using Raycasting and Riemannian Motion Policies at kHz rates for MAVs Open
In this paper, we present a novel method for using Riemannian Motion Policies on volumetric maps, shown in the example of obstacle avoidance for Micro Aerial Vehicles (MAVs). While sampling or optimization-based planners are widely used fo…
View article: A Framework for Collaborative Multi-Robot Mapping using Spectral Graph Wavelets
A Framework for Collaborative Multi-Robot Mapping using Spectral Graph Wavelets Open
The exploration of large-scale unknown environments can benefit from the deployment of multiple robots for collaborative mapping. Each robot explores a section of the environment and communicates onboard pose estimates and maps to a centra…
View article: SphNet: A Spherical Network for Semantic Pointcloud Segmentation
SphNet: A Spherical Network for Semantic Pointcloud Segmentation Open
Semantic segmentation for robotic systems can enable a wide range of applications, from self-driving cars and augmented reality systems to domestic robots. We argue that a spherical representation is a natural one for egocentric pointcloud…
View article: Baking in the Feature: Accelerating Volumetric Segmentation by Rendering Feature Maps
Baking in the Feature: Accelerating Volumetric Segmentation by Rendering Feature Maps Open
Methods have recently been proposed that densely segment 3D volumes into classes using only color images and expert supervision in the form of sparse semantically annotated pixels. While impressive, these methods still require a relatively…
View article: Learning Agent-Aware Affordances for Closed-Loop Interaction with Articulated Objects
Learning Agent-Aware Affordances for Closed-Loop Interaction with Articulated Objects Open
Interactions with articulated objects are a challenging but important task for mobile robots. To tackle this challenge, we propose a novel closed-loop control pipeline, which integrates manipulation priors from affordance estimation with s…