Not only equipped with basic functions, such as auto-pilot with obstacles avoiding features, visual monitoring and audio communication, the robotic solution is also expected to be equipped with artificial intelligence. The backend real-time monitoring system can be centralised at one place without involving the stationing staff in closely viewing all the screens all the time as the AI-enabled video(and sound)-analytics technology would help pre-analyse the motions (& the sounds) observed from the person(s)-in-custody (PICs).

The robot itself must be able to :

• get through a minimum width of passage as narrow as 0.8m;
• silently cruise over bare concrete floor without inducing unacceptable noise (not continuously higher than 45 dBA);
• have a wide-angle aperture of the video-camera(s) for viewing but without much distorting the aspect ratio;
• real-time transmission and storing of the video footage to the central station without any delay or intermission;
• equipped with multiple (and movable) camera(s) to allow the user to observe the prison cell on a side of the robot as well as observe the environment in front and surroundings.

Owing to the high security concern to the CSD, no access to any data is allowed via the extranet, a cloud service or whatever external means.

• Design & Development – 1 to 2 months
• Offsite/Pre-delivery Trial Run – 1 month
• Onsite Trial Run – 1 to 2 months