- 255
- 1 217 290
aslteam
Приєднався 3 лис 2008
Our mission and dedication is to create robots and intelligent systems that are able to autonomously operate in complex and diverse environments. We are interested in the mechatronic design and control of systems that autonomously adapt to different situations and cope with our uncertain and dynamic daily environment. We are fascinated by novel robot concepts that are best adapted for acting on the ground, in the air and in the water. We are furthermore keen to give them the intelligence to autonomously navigate in challenging environments. This includes novel methods and tools for perception, abstraction, mapping and path planning.
Task Adaptation in Industrial Human-Robot Interaction: Leveraging Riemannian Motion Policies
This is the accompanying video of our RSS 2024 paper titled "Task Adaptation in Industrial Human-Robot Interaction: Leveraging Riemannian Motion Policies".
Abstract:
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 can adjust their motion based on human intentions, enabling dynamic task planning and adaptation. Addressing the needs of industrial applications, we propose a motion control framework that (i) removes the need for manual control of the robot’s movement; (ii) facilitates the formulation and combination of complex tasks; and (iii) allows the seamless integration of human intent recognition and robot motion planning. For this purpose, we leverage a modular and purely reactive approach for task parametrization and motion generation, embodied by Riemannian Motion Policies. The effectiveness of our method is demonstrated, evaluated, and compared to a representative state-of-the-art approach in experimental scenarios inspired by realistic industrial Human-Robot Interaction settings.
Reference:
Mike Allenspach, Michael Pantic, Rik Girod, Lionel Ott, Roland Siegwart; "Task Adaptation in Industrial Human-Robot Interaction: Leveraging Riemannian Motion Policies"; Robotics, Science and Systems (RSS) 2024
Affiliations:
All authors are with the Autonomous Systems Lab, ETH Zurich, 8092 Switzerland.
Abstract:
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 can adjust their motion based on human intentions, enabling dynamic task planning and adaptation. Addressing the needs of industrial applications, we propose a motion control framework that (i) removes the need for manual control of the robot’s movement; (ii) facilitates the formulation and combination of complex tasks; and (iii) allows the seamless integration of human intent recognition and robot motion planning. For this purpose, we leverage a modular and purely reactive approach for task parametrization and motion generation, embodied by Riemannian Motion Policies. The effectiveness of our method is demonstrated, evaluated, and compared to a representative state-of-the-art approach in experimental scenarios inspired by realistic industrial Human-Robot Interaction settings.
Reference:
Mike Allenspach, Michael Pantic, Rik Girod, Lionel Ott, Roland Siegwart; "Task Adaptation in Industrial Human-Robot Interaction: Leveraging Riemannian Motion Policies"; Robotics, Science and Systems (RSS) 2024
Affiliations:
All authors are with the Autonomous Systems Lab, ETH Zurich, 8092 Switzerland.
Переглядів: 155
Відео
Safe Low-Altitude Navigation in Steep Terrain with Fixed-Wing Aerial Vehicles
Переглядів 205Місяць тому
Fixed-wing aerial vehicles provide an efficient way to navigate long distances or cover large areas for environmental monitoring applications. By design, they also require large open spaces due to limited maneuverability. However, strict regulatory and safety altitude limits constrain the available space. Especially in complex, confined, or steep terrain, ensuring the vehicle does not enter \ia...
Watching the Air Rise: Learning-Based Single-Frame Schlieren Detection
Переглядів 266Місяць тому
This is the accompanying video of our 2024 IEEE International Conference on Robotics and Automation (ICRA) paper "Watching the Air Rise: Learning-Based Single-Frame Schlieren Detection". Abstract: Detecting air flows caused by phenomena such as heat convection is valuable in multiple scenarios, including leak identification and locating thermal updrafts for extending UAV flight duration. Unfort...
WindSeer: Real-time volumetric wind prediction over complex terrain aboard a small UAV
Переглядів 191Місяць тому
This is the accompanying video of our 2024 Nature Communications paper "Real-time volumetric wind prediction over complex terrain aboard a small UAV". Abstract: Real-time high-resolution wind predictions are beneficial for various applications including safe crewed and uncrewed aviation. Current weather models require too much compute and lack the necessary predictive capabilities as they are v...
SC-Explorer: Incremental 3D Scene Completion for Safe and Efficient Exploration Mapping and Planning
Переглядів 542Місяць тому
Paper: arxiv.org/abs/2208.08307 Code: github.com/ethz-asl/ssc_exploration Abstract: Exploration of unknown environments is a fundamental problem in robotics and an essential component in numerous applications of autonomous systems. A major challenge in exploring unknown environments is that the robot has to plan with the limited information available at each time step. While most current approa...
COIN-LIO: Complementary Intensity-Augmented LiDAR Inertial Odometry (ICRA 2024)
Переглядів 776Місяць тому
We present COIN-LIO, a LiDAR Inertial Odometry pipeline that tightly couples information from LiDAR intensity with geometry-based point cloud registration. The focus of our work is to improve the robustness of LiDAR-inertial odometry in geometrically degenerate scenarios, like tunnels or flat fields. We project LiDAR intensity returns into an intensity image, and propose an image processing pip...
Geranos: a Novel Tilted-Rotors Aerial Robot for the Transportation of Poles
Переглядів 1,3 тис.3 місяці тому
Geranos: a Novel Tilted-Rotors Aerial Robot for the Transportation of Poles Authors: Nicolas Gorlo, Samuel Bamert, Rafael Cathomen, Gabriel Käppeli, Mario Müller, Tim Reinhart, Henriette Stadler, Hua Shen, Eugenio Cuniato, Marco Tognon, Roland Siegwart In challenging terrains, constructing structures such as antennas and cable-car masts often requires the use of helicopters to transport loads v...
Panoptic Vision-Language Feature Fields
Переглядів 3445 місяців тому
This is the accompanying video of our IEEE RA-L 2024 paper "Panoptic Vision-Language Feature Fields". Paper: doi.org/10.1109/LRA.2024.3354624 arXiv: arxiv.org/abs/2309.05448 Project page: ethz-asl.github.io/pvlff Code: github.com/ethz-asl/pvlff Abstract: Recently, methods have been proposed for 3D open-vocabulary semantic segmentation. Such methods are able to segment scenes into arbitrary clas...
Safe Low-Altitude Navigation in Steep Terrain with Fixed-Wing Aerial Vehicles
Переглядів 1,4 тис.5 місяців тому
Abstract Fixed-wing aerial vehicles provide an efficient way to navigate long distances or cover large areas for environmental monitoring applications. By design, they also require large open spaces due to limited maneuverability. However, strict regulatory and safety altitude limits constrain the available space. Especially in complex, confined, or steep terrain, ensuring the vehicle does not ...
Autonomous Christmas Lab 2023
Переглядів 1,4 тис.6 місяців тому
Wishing you and your loved ones merry Christmas, happy holidays, and a happy New Year from everyone at the Autonomous Systems Lab at ETH Zürich!
ISAR: A Benchmark for Single- and Few-Shot Object Instance Segmentation and Re-Identification
Переглядів 3107 місяців тому
This is the accompanying video of our WACV 2024 paper "ISAR: A Benchmark for Single- and Few-Shot Object Instance Segmentation and Re-Identification". Paper: arxiv.org/abs/2311.02734 Project page: nicogorlo.github.io/isar_wacv24 Code: github.com/nicogorlo/isar Abstract: Most object-level mapping systems in use today make use of an upstream learned object instance segmentation model. If we want ...
A perching and drilling aerial robot
Переглядів 2,3 тис.8 місяців тому
This video presents a novel perching and tilting aerial robot for precise and versatile power tool work on vertical walls. The system was developed as part of the AITHON ETH Zürich Bachelor student focus project and presented at IEEE IROS 2023. It combines a compact integrated perching drone design with a concrete drill's heavy payload and reaction forces. Winner of IROS 2023 Best Paper Award o...
Reinforcement Learning for Outdoor Balloon Navigation - Outdoor Tests
Переглядів 1,6 тис.9 місяців тому
Videos of the outdoor tests conducted as part of the "Reinforcement Learning for Outdoor Balloon Navigation" published in IEEE-RAM
Multi-directional Interaction Force Control with an Aerial Manipulator under External Disturbances
Переглядів 68510 місяців тому
This is the accompanying video of our Springer Autonomous Robots paper "Multi-directional Interaction Force Control with an Aerial Manipulator under External Disturbances". Abstract: To improve accuracy and robustness of interactive aerial robots, the knowledge of the forces acting on the platform is of uttermost importance. The robot should distinguish interaction forces from external disturba...
Learning to Open Doors with an Aerial Manipulator
Переглядів 1,3 тис.10 місяців тому
This is the accompanying video of our IROS2023 paper "Learning to Open Doors with an Aerial Manipulator". Abstract: The field of aerial manipulation has seen rapid advances, transitioning from push-and-slide tasks to interaction with articulated objects. The motion trajectory of these complex actions is usually hand-crafted or a result of online optimization methods like Model Predictive Contro...
[CVPR 2023] Unsupervised Continual Semantic Adaptation through Neural Rendering
Переглядів 1,4 тис.Рік тому
[CVPR 2023] Unsupervised Continual Semantic Adaptation through Neural Rendering
Continual Adaptation of Semantic Segmentation Using Complementary 2D-3D Data Representations
Переглядів 256Рік тому
Continual Adaptation of Semantic Segmentation Using Complementary 2D-3D Data Representations
Design and Control of a Micro Overactuated Aerial Robot with an Origami Delta Manipulator
Переглядів 1,3 тис.Рік тому
Design and Control of a Micro Overactuated Aerial Robot with an Origami Delta Manipulator
Dynablox: Real-time Detection of Diverse Dynamic Objects in Complex Environments
Переглядів 3,8 тис.Рік тому
Dynablox: Real-time Detection of Diverse Dynamic Objects in Complex Environments
Mixed Reality Human-Robot Interface for 6DoF Trajectory Planning of Omnidirectional Aerial Vehicles
Переглядів 805Рік тому
Mixed Reality Human-Robot Interface for 6DoF Trajectory Planning of Omnidirectional Aerial Vehicles
Resilient Terrain Navigation with a 5 DOF Metal Detector Drone (ICRA 2023)
Переглядів 8 тис.Рік тому
Resilient Terrain Navigation with a 5 DOF Metal Detector Drone (ICRA 2023)
Robust Sampling-based Control of Mobile Manipulators for Interaction with Articulated Objects
Переглядів 1,4 тис.Рік тому
Robust Sampling-based Control of Mobile Manipulators for Interaction with Articulated Objects
Continual Adaptation of Semantic Segmentation using Complementary 2D-3D Data Representations
Переглядів 585Рік тому
Continual Adaptation of Semantic Segmentation using Complementary 2D-3D Data Representations
A Planning-and-Control Framework for Aerial Manipulation of Articulated Objects
Переглядів 1,9 тис.Рік тому
A Planning-and-Control Framework for Aerial Manipulation of Articulated Objects
Adaptive Tank-based Control for Aerial Physical Interaction with Uncertain Dynamic Environments
Переглядів 576Рік тому
Adaptive Tank-based Control for Aerial Physical Interaction with Uncertain Dynamic Environments
Under the Sand: A UAV with Ground Penetrating Synthetic Aperture Radar
Переглядів 6 тис.Рік тому
Under the Sand: A UAV with Ground Penetrating Synthetic Aperture Radar
Closed-Loop Next-Best-View Planning for Target-Driven Grasping
Переглядів 1,5 тис.Рік тому
Closed-Loop Next-Best-View Planning for Target-Driven Grasping
Fast and Compute-efficient Sampling-based Local Exploration Planning via Distribution Learning
Переглядів 852Рік тому
Fast and Compute-efficient Sampling-based Local Exploration Planning via Distribution Learning
Avalmapper: Monitoring avalanches with long endurance UAVs
Переглядів 1,1 тис.2 роки тому
Avalmapper: Monitoring avalanches with long endurance UAVs
Very impressive result! Thanks for open-sourcing your works.
Looks interesting, but there is no code on github...
Will be soon!
Can I ask what method you use to supply power to the YUMI robot? I don't see you using an AC power cord
I'm still having Yaw issues 😢
Keep your seat belts tight. The wind flows like water between the rocks.
@1:30 is that a bat? 🙂
does voxblox work on VOXL 2 board with ROS ??
Great Reseach
Thank you for sharing the video showcasing the implementation validation and presentation. I greatly appreciate your effort in providing us with valuable insights. just I think It would be better for viewers to access the link to the research paper on the "video description box" directly.
An amazing research project.
Which drone have you used for this experiment?
It's a DJI M600 Pro. It has a good payload and works very reliable. However, the custom ground penetrating radar payload is the core contribution of this research project. So essentially any drone could be used.
I came here after reading the paper.. Interesting work 🎉
is the code for this work available?
same question
Im guessing these guys are a bit confused as to why a bunch of views and comments just showed up 6 years later. Cool machine, just dont let it run on human souls, bad things will happen!
Out of curiosity, does this machine happen to be powered by human souls?
Looks like these scientists WERE careful with their Command
Sir can u plz tell me how can we represent the carriage width, formation width, and total width of a road by using the segment mapping method?
I hev pi 4 with 4gb rem is is enough? thx end kisses
I'm sure ilness can gather the rest
Awesome.
Amazing 👍
I like the idea however what is a useful applicationbfor this?
Great procedure! and ecelllent simulator! Thanks for making it open source
Excellent
what kind of drone do you use?
Hummingbird quadrotor from Ascending Technologies! it is written in section 5 of their paper.
Hanzhen harmonic drive gear , strain wave reducer, robot joint , over 30 years experience
how does it differs from regular depth point clouds?
Hi! can you please tell me which kind of lidar is installed on the robot-car?
That's genius
This is a great experiment. makes a lot of sense to leverage this local/global hierarchy, and this experiment provides us a wonderful proof point.
Wow! Cool😀
Amazing!
So fascinating to see whats possible with drones. Next Project: Put a Softgripper at the end to pick blackberries and other Fruits gently! (And send it to me :D here are lots of wild Blackberries i cant reach by hand)
You might be interested in the drone we've built ;) Check out our videos for a fully omnidirectional OMAV equipped with a soft adaptive gripper! Our drone could surely grasp those hard to reach Blackberries!
what are the real life objectives of this uav?
Clear message, clear structure, easy to understand, thank you
what does OMAV stand for?
Amazing work! How did you guys handle the noise in sensors (Gyro/Accelerometer) ?
kalman filtering based techniques are used to achieve this i guess...also as far as the localization problem is concerned, the drone has reflective markers so im assuming a camera capturing mechanism is present for 3d localization.
@@harshavardhankulkarni6599 yep! Makes sense
Nice work!
Is this running a MAVSDK based flight platform?
smoot af
This is really cool! An area of interest of mine is that of E-Bike riding robots and this technology could be very applicable to that.
This is really a cool capability and I can see many application where it could be quite beneficial!
This is really neat! Creating segments be beneficial for many applications!
This is cool! Collaboration is typically challenging for optimization algorithms.
I can see these being integrated with jet packs such as those developed by Gravity Industries to provide mutual support.
Hanzhen harmonic gear , robot gear reducer , over 30 years experience
Our manufacturer make magnetic wheels
What is the format of 2D/3D Building info? do you get it from BIM or revit ? or something else?
great！！！
May I know the software for programming and computer vision? now I working on my final year project about this topic...Thank You..