# ICRA Research
# Presentation [[Ref](https://2024.ieee-icra.org/presentation-instruction/)]
* **10 minutes** (7 minutes for your presentation and 3 minutes for Q&A)
* Please **bring and connect own laptop computer** to the LCD projector for your presentation.
* Please make sure your computer is connected to the LCD projector in the **speaker preview room before using**. Session rooms are also available throughout the day and you can check your presentation in advance at the venue.
* Speaker Preview Room Opening Hours
* **Monday 13th May – 15:00-18:00**
* Tuesday 14th May – 08:30-18:00
* Wednesday 15th May – 08:30-18:00
* Thursday 16th May – 08:30-16:30
* **10 minutes before the start of the session**


# Poster
* Please have your poster on the board **at least 5 minutes** before your session. We will provide material with which to attach your poster to the board.
* Each session lasts 90 minutes.
* No power supplies or displays for video presentations.

# Program
See pdf from calendar or pdf.

# Workshops
## Mon. 13th May
### Workshop on Field Robotics
- https://norlab-ulaval.github.io/workshop_field_robotics_icra2024/
- UAV Path Planning for Agricultural Applications, Marija Popović
(University of Bonn) [11:10 - 11:30]
- **Robust navigation in degraded environments in air and on land, Sebastian Scherer
(Carnegie Mellon University) [11:30 - 11:50]**
- **Overview of the Last 20 Years in Field Robotics, Sanjiv Singh
(Carnegie Mellon University) [13:30 - 14:30]**
- Collaborative AI robots for adaptation of diverse environments and innovation of infrastructure construction, Keiji Nagatani
(The University of Tokyo) [16:45 - 17:05]

### 2nd HERMES Workshop: Multi-Robot Sensing & Perception in Extreme Environments
- https://hermes-workshop.com/2024.html
- **Anchor-oriented Multi-Robot Coverage without Global Localization, Aiman Munir et al. (University of Georgia) [11:50]**
- Distributed multi-robot mapping in the real world: lessons learned and future opportunities, Luca Carlone (MIT) [13:35]

### 3rd Workshop on Future of Construction: Lifelong Learning Robots in Changing Construction Sites
- https://construction-robots.github.io/
- **Margarita Chli - ETH Zurich [14:20]**

### Dynamic Duos: Human-Robot Co-Worker Adaptation in Manufacturing
- www.rocoadapt.eu
- Navigating the Upward Spiral for Sensorimotor Learning, Shuran Song (stanford) [09:20 AM - 09:40 AM]

### Back to the Future: Robot Learning Going Probabilistic
- https://probabilisticrobotics.github.io/
- **Safe and Efficient Exploration in Bayesian Model-based Reinforcement Learning, Andreas Krause [09:00-09:30]**
- End-to-end yet Modular Probabilistic Robot Learning, Peter Karkus (nvidia) [11:30-12:00]
- From Words to Actions, Andy Zeng (deepmind) [15:30-16:00]

### BIMANUAL MANIPULATION: ON KITCHEN CHALLENGES
- https://sites.google.com/view/icra2024-ws-bi-mp/home
- **Neuro-Symbolic Concepts for Robotic Manipulation, Jiayuan Mao (MIT) [09:45 - 10:15]**

### Tutorial on Ergodic Control
- https://ergodiccontrol.github.io/
- Applications: Search with UAVs, (Stefan) [13:20-13:40]
- Applications: Duration, reachability and optimality in exploration, (Ian, Henry) [13:40-14:00]
- Applications: Runtime robot learning of neural networks for perception and RL, (Todd) [14:20-14:40]
- **Multi-Robot Coverage, (Guillaume) [16:00-16:20]**

### [tutorial] Connected Robotics Platform for ROS Deployment in Real-world Network Settings
- https://github.com/5G-ERA/docs/tree/main/Workshops/ICRA24
- Join any session

## Fri. 17th May
### Advancements in Trajectory Optimization and Model Predictive Control for Legged Systems – 2nd Edition
- https://atompc-workshop.github.io/
- Trajectory optimization with MuJoCo, Yuval Tassa (deepmind) [14:30 - 15:00]

### Robot Software Architectures (RSA24)
- https://rsa24.weebly.com/
- Boston dynamic ai [10:45-11:15]
- ros2 evolution [11:45-12:15]

### First Workshop on Vision-Language Models for Navigation and Manipulation
- https://vlmnm-workshop.github.io/
- LLM-based Task and Motion Planning for Robots, Chuchu Fan (mit) [9:20 - 9:40]
- Language as Bridge for Sim2Real, Roberto Martín-Martín (University of Texas at Austin) [13:30 - 13:50]
- Mobile Manipulation, Multi-Agent Coordination, Long Horizon Tasks, Jeannette Bohg (Stanford University) [14:10 - 14:30]

### Breaking Swarm Stereotypes
- https://hauertlab.com/breaking-swarm-stereotypes-workshop-icra-2024/
- Switching between decentralised and centralised control in swarm robotics, [14:40-15:10]
- Intelligent Multi Agents for Exploration and Settlement in Unknown and Unexplored Areas, Yasuharu Kunii [15:45-16:15]

### 4th Workshop on Representing and Manipulating Deformable Objects
- https://deformable-workshop.github.io/icra2024/
- TBA, Dmitry Berenson (University of Michigan) [16:15- 16:45]

### How to manage fleets of robots with open source software
- https://www.theconstructsim.com/icra-tutorial-how-to-manage-fleets-of-robots-with-open-source-software/

# Topics
## [Learning to walk in confined spaces using 3D representation](https://takahiromiki.com/publication-posts/learning-to-walk-in-confined-spaces-using-3d-representation/?i=1)

Legged robots have the potential to traverse complex terrain and access confined spaces beyond the reach of traditional platforms thanks to their ability to carefully select footholds and flexibly adapt their body posture while walking. However, robust deployment in real-world applications is still an open challenge. In this paper, we present a method for legged locomotion control using reinforcement learning and 3D volumetric representations to enable robust and versatile locomotion in confined and unstructured environments. By employing a two-layer hierarchical policy structure, we exploit the capabilities of a highly robust low-level policy to follow 6D commands and a high-level policy to enable three-dimensional spatial awareness for navigating under overhanging obstacles. Our study includes the development of a procedural terrain generator to create diverse training environments. We present a series of experimental evaluations in both simulation and real-world settings, demonstrating the effectiveness of our approach in controlling a quadruped robot in confined, rough terrain. By achieving this, our work extends the applicability of legged robots to a broader range of scenarios.