Vision system for teat detection

A vision system comprising:

• a robotic arm that attaches a teat cup to a dairy livestock in a stall;
• a laser that generates a plurality of profile signals, each profile signal comprising information associated with a relative distance between the laser and at least a portion of the dairy livestock;
• one or more memory devices that stores historical teat location information for a plurality of teats of the dairy livestock;
• and
• a processor communicatively coupled to the laser and the one or more memory devices, wherein the processor:
• determines, from the historical teat location information, a first expected teat position associated with a first teat and a second expected teat position associated with a second teat;
• determines a distance between the first expected teat position and the second expected teat position;
• determines that the distance is less than or equal to a predetermined distance;
• in response to determining that the distance is less than or equal to the predetermined distance, commands the robotic arm to move to a first location that is between the first expected teat position and the second expected teat position;
• commands the robotic arm to move up a predetermined amount after the robotic arm moves to the first location;
• commands the laser to perform a scan of the dairy livestock after the robotic arm moves up the predetermined amount;
• accesses a first profile signal generated by the laser from the first scan;
• determines that both first and second teats are found in the first profile signal;
• and
• in response to determining that both first and second teats are found in the first profile signal, commands the robotic arm to attach the teat cup to the first or second teats.