We need to move our robot in a simple warehouse environment using SLAM and AI based perception. I am attaching the video of the environment.
1. Our input can be in the form of video stream from one or two cameras.
2. The algorithm shall be able to provide current location of the robot using SLAM.
3. The AI based perception shall provide current obstacle map.
4. Preferably, the algorithm should be able to decide path from Point A in SLAM generated map to Point B.
5. The output can be in the form of forward, reverse, left, right signal or motion vector sequence for next few steps.
Coding language shall be python. Preferred tool: Tensor flow.