- 44
- 10 292
Aamir Ahmad
Приєднався 27 лис 2007
Enjoy videos of our robots and experiments from the Flight Robotics and Perception Group, which I am currently leading. For more details about my group, visit aamirahmad.de
Multi-Task Reinforcement Learning in Continuous Control
Paper title: Multi-Task Reinforcement Learning in Continuous Control with Successor Feature-Based Concurrent Composition
Authors: Yu Tang Liu and Aamir Ahmad
Conference: The European Control Conference (ECC) 2024
Paper download link (arxiv version): arxiv.org/abs/2303.13935
Open Source Code: github.com/robot-perception-group/concurrent_composition
Authors: Yu Tang Liu and Aamir Ahmad
Conference: The European Control Conference (ECC) 2024
Paper download link (arxiv version): arxiv.org/abs/2303.13935
Open Source Code: github.com/robot-perception-group/concurrent_composition
Переглядів: 18
Відео
Autonomous Airship Overflying Przewalski's Horses in Hungary
Переглядів 563 місяці тому
Autonomous flight test of an airship overflying przewalski's horses at the Hortobagy National Park in Hungary.
Autonomous Airship Flight Test at HNP, Hungary
Переглядів 2623 місяці тому
Autonomous airship flight test at the Hortobagy National Park (HNP), Hungary.
MPC-based Formation Control Test with a Single Airship
Переглядів 1033 місяці тому
Flight test with a single airship running a percpetion-driven formation controller, assuming a static target at the origin.
Perception-optimal Airships Formation Control with Obstacle Avoidance.
Переглядів 153 місяці тому
Perception-optimal autonomous formation control of airships tracking a moving person in an environment with static obstacles.
Perception-optimal Airships Formation Control
Переглядів 63 місяці тому
Perception-optimal Autonomous Formation Control of Airships Tracking a Moving Person
Reinforcement Learning for Airship Control: Yu Tang Liu's 2023 Year in Review at FRPG
Переглядів 917 місяців тому
Reinforcement Learning for Airship Control: Yu Tang Liu's 2023 Year in Review at FRPG
Autonomous Bird Tracking - Systems and Methods: Nico Klar's 2023 Year in Review at FRPG
Переглядів 237 місяців тому
Autonomous Bird Tracking - Systems and Methods: Nico Klar's 2023 Year in Review at FRPG
Synthetic Data Generation, Animal Pose and Dynamic SLAM: Elia Bonetto's 2023 Year in Review at FRPG
Переглядів 357 місяців тому
Synthetic Data Generation, Animal Pose and Dynamic SLAM: Elia Bonetto's 2023 Year in Review at FRPG
Autonomous Airships, Formations and Tracking Animals: Eric Price' 2023 Year in Review at FRPG
Переглядів 3887 місяців тому
Autonomous Airships, Formations and Tracking Animals: Eric Price' 2023 Year in Review at FRPG
Animal Behavior Inference from Drone Videos: Pranav Khandelwal's 2023 Year in Review at FRPG
Переглядів 667 місяців тому
Animal Behavior Inference from Drone Videos: Pranav Khandelwal's 2023 Year in Review at FRPG
FRPG's Autonomous Blimp detects, tracks and follows a group of Przewalski's horses
Переглядів 20411 місяців тому
Our #airship visually tracks and #autonomously follow Przewalski's horses roaming free in the Steppe. In the video, we show how the blimp 'fights' the wind to keep the horses centered in its field of view. Flight Robotics and Perception Group's (FRPG) endeavor to scale up #biodiversity #monitoring, both spatially and temporally, through #intelligent #aerial #robots is a step closer to reality! ...
Animal Behavior Inference Framework based on Smarter-Labelme
Переглядів 13111 місяців тому
Title: A Framework for Fast, Large-scale, Semi-Automatic Inference of Animal Behavior from Monocular Videos Authors: Eric Price, Pranav C. Khandelwal, Daniel I. Rubenstein and Aamir Ahmad Abstract: An automatic, quick, accurate, and scalable method for animal behavior inference using only videos of animals offers unprecedented opportunities to understand complex biological phenomena and answer ...
Synthetic Data-based Detection of Zebras in Drone Imagery
Переглядів 70Рік тому
Our bleeding-edge work on generating synthetic zebras in wild environments and how we are able to detect real zebras using only that synthetic data! Thanks to Elia Bonetto for a great leap in synthetic data related research thread of our Flight Robotics and Perception Group. Data: keeper.mpdl.mpg.de/d/12abb3bb6b12491480d5/ Code: github.com/eliabntt/GRADE-RR Paper: arxiv.org/abs/2305.00432
Viewpoint-driven Formation Control of Airships for Cooperative Target Tracking
Переглядів 152Рік тому
Journal: IEEE Robotics and Automation Letters (RA-L) 2023 (accepted, DOI awaited) Paper Title: Viewpoint-driven Formation Control of Airships for Cooperative Target Tracking Authors: Eric Price, Michael J. Black and Aamir Ahmad Abstract: For tracking and motion capture (MoCap) of animals in their natural habitat, a formation of safe and silent aerial platforms, such as airships with on-board ca...
GRADE: Generating Realistic Animated Dynamic Environments for Robotics Research
Переглядів 272Рік тому
GRADE: Generating Realistic Animated Dynamic Environments for Robotics Research
Title: SmartMocap: Joint Estimation of Human and Camera Motion using Uncalibrated RGB Cameras
Переглядів 172Рік тому
Title: SmartMocap: Joint Estimation of Human and Camera Motion using Uncalibrated RGB Cameras
Autonomous soaring with end-to-end reinforcement learning
Переглядів 443Рік тому
Autonomous soaring with end-to-end reinforcement learning
Active visual SLAM from static to dynamic environments
Переглядів 95Рік тому
Active visual SLAM from static to dynamic environments
Autonomous blimp navigation with deep residual reinforcement learning
Переглядів 52Рік тому
Autonomous blimp navigation with deep residual reinforcement learning
NMPC based Perception-driven formation control of autonomous blimps and more
Переглядів 45Рік тому
NMPC based Perception-driven formation control of autonomous blimps and more
Reinforcement learning based autonomous multi-rotor landing on moving platforms
Переглядів 68Рік тому
Reinforcement learning based autonomous multi-rotor landing on moving platforms
SmartMocap: Joint Estimation of Human and Camera Motion using Uncalibrated RGB Cameras
Переглядів 548Рік тому
SmartMocap: Joint Estimation of Human and Camera Motion using Uncalibrated RGB Cameras
Perception-driven Formation Control of Airships
Переглядів 77Рік тому
Perception-driven Formation Control of Airships
Deep Residual Reinforcement Learning based Autonomous Blimp Control (IROS 2022)
Переглядів 1712 роки тому
Deep Residual Reinforcement Learning based Autonomous Blimp Control (IROS 2022)
Deep Residual Reinforcement Learning based Autonomous Blimp Control
Переглядів 1062 роки тому
Deep Residual Reinforcement Learning based Autonomous Blimp Control
AirPose: Multi-View Fusion Network for Aerial 3D Human Pose and Shape Estimation
Переглядів 6372 роки тому
AirPose: Multi-View Fusion Network for Aerial 3D Human Pose and Shape Estimation
Active Visual SLAM with Independently Rotating Camera (ECMR 2021)
Переглядів 2082 роки тому
Active Visual SLAM with Independently Rotating Camera (ECMR 2021)
Tolle Arbeit!
More links with details please.
Assalamoalaikum sir I am looking for a postdoc position in robotics. my expertise is in the gait generation of the humanoid robot on complex terrains. My PhD from MANIT Bhopal is on the verge of completion in the next six months.
Love this, great work. As an FPV pilot I think everyone over-hypes quadcopter style drone technology without understanding it's limitations and simpler things like lighter than air designs get overlooked. 1 hr of flight time is awesome and has so many applications. Keep it up.
Very interesting! Is the learned controller strictly for the thermaling portion of flight or also responsible for the "search" flight pattern between thermals?
Aamir. how breathtaking view,
Hi, I'd love to test this in my next project but I have no coding skills, how can I mail you about it? :)
Hi Aamir, thanks for the great video! Is there any sample of the quality that can be achieved with your method and 4-5 cameras? And the cameras can be of any type, right? I have so many questions :) Thank you!
Yes, the cameras can be of any type. Some results using 7 fixed cameras are shown at 2:07. You'll get a similar quality with 4-5 cameras.
@@nitinsaini1991 thank you so much! Is it free to try then? And is there any precompiled .exe available?
@@giacomospaconi7189 sorry for the late reply. You can try our method for free: github.com/robot-perception-group/SmartMocap
From kashmir 🖐🖐🖐🖐
Hi, sir I need the help of yours on making robots for my college project, I have to make 4 robots that are controlled by 1 external placed camera module. In this project No sensors for navigation/object detection to be mounted on the robot. Only 1 external placed camera module will do Live Image Processing and control the robots. I need your help and valuable suggestions on this, I am planning to use PID Control, Model predictive control(MPC). (hardware- NVIDIA jetson nano for image processing)
Beautiful photography of an historic event!
Wow! Congrats! The blimp looks so cool against the snowy background.
Congratulations!
Brilliant performance...May Allah bless you with more success and prosperity.. Inshallah..
very cool