This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

IWR6843ISK: Human detection on a mobile robot: Detect human presence/motion while the sensor is moving

Part Number: IWR6843ISK

Hello,

We are looking for human presence sensors for our indoor mobile robotics application. The robot needs to stop operation when a human motion/presence is detected within 5 meters in an enclosed space.

The sensor will be mounted on a mobile robot. We need to just know a single detection value (Yes/No) of human in the vicinity.

Can the mmwave radars provide this functionality ? Distinguishing human presence from other objects in an enclosed area.

Thanks,

Aswin

  • Hello Aswin,

    Just so I understand, you would like to detect only humans to stop the robot? Or you would like to stop the robot for any object? 

    If you would like to stop the robot for any objects, then we have a demo that does this. Please see the Safety Bubble demo in our Industrial Toolbox on dev.ti.com/tirex. The demo shows this functionality using a Turtlebot and the ROS platform.  

  • I would like to detect only humans to stop the robot.

    We have other sensors on the robot to detect and avoid other obstacle.

  • This is a challenging problem. We have demos where the radar is able to track moving people (such as people counting and area scanner), but the sensor is stationary. For tracking while the sensor is moving, the radar has to know the current robot velocity and offset the current velocity with the velocity of the tracked objects. We don't have any demo that has solved this today. 

    I would suggest experimented with our two demos that utilize tracking (area scanner and people counting) on your robot as is. This will show the capabilities of the tracking while moving. However, because everything is moving relative to the radar, the tracking algorithm will track everything, not just humans. In combination with a vision sensor, this idea becomes more feasible. 

  • Thank you very much for the clarification. We will look into these possibilities.