This page serves as a summary of RAICAM's first workshop. It provides an overview of the skills brought to the table by each member of the team and a collection of possible scenarios that were brainstormed during the meeting. These scenarios represent potential avenues for future exploration and collaboration within the RAICAM team.

Untitled

☑️ Tasks

Untitled

• Hamidreza video:

Videos - HRII - IIT

• Alperen: Digital Twin Paper (removed for confidentiality)

List of possible scenarios of collaboration:

What ideas should we discuss at Friday's meeting? - Add the description and list of steps of the scenario

Changda:

1. Emergency Response and Hazardous Material Handling: Develop scenarios where robots perform emergency interventions in environments hazardous to humans, such as chemical spills or radiation leaks. The robots could find a path to the hazard position very fast and then use manipulation tools to contain the hazard or collect samples for analysis.
   1. Need: Slam, Path planning, Manipulation.
   2. Judgment criteria: Total reaction speed from finding the hazard to taking control of it.
2. Search and Rescue Operations in Collapsed Structures: Design missions where robots navigate through simulated collapsed buildings to locate and rescue trapped individuals. This could involve manipulating debris to clear paths and using sensors or drones to detect signs of life. Though we may not have legged robot, we can still do it with IIT’s mobile robot manipulation platform.
   1. Need: Navigation of robots and drones, Path planning, Manipulation, Multi-robot cooperation.
   2. Judgment criteria: Total reaction speed of rescue. Cooperation among different robots.

Alperen:

I think we can start with a simpler project for the sprint demo, we might try to implement similar task scenarios as in the paper with two robot collaboration with a mobile land robot(legged or wheeled) and an aerial robot (quadcopter or flap wing). The scenarios were:

Scenario 1: Freely controlling the robots

The operator can observe the mobile robots and explore and interact with the software freely to get used to the controlling of robots given the interface.

Scenario 2: Inspection and Mapping

Real-time teleoperating with the robot team to inspect the task environment to discover dark corners. If we have an appropriate sensor on the land robot, we might also try point cloud mapping of the environment

Scenario 3: Hazard identification

In the paper, they have a robot with a special sensor to measure radioactive contamination and create radiation mapping. We can go with Trajectory optimization data driven approaches (e.g., deep learning or something much simpler like placing risk sign symbols randomly around the lab, implementing image processing and trying to visualize and identify potential hazards and mark them on the map.

An alternative to this task is if we have a manipulator arm on the mobile land robot, we can place an item such as a red ball inside the lab whose location is unknown, and the operator can try to pick up the item and carry it to somewhere else. Or we might even try a swabbing task.

Sasanka:

Semi-autonomous multi-agent swabbing