The robotic machine which was developed by the team led by Dr. Chao Chen in Department of Mechanical and Aerospace Engineering, Monash University, is capable of identifying apple fruits, picking and depositing them in harvest container in as little as seven second when working on full capacity.
It was reported that the field trial of the robot at Fankhauser Apples, Drouin in Victoria shows it having a eight-five (85) percent success rate with an average of nine second per apple picked and deposited working at half of it capacity and with just about 6% of all harvested apple damaged due to stem removal.
Read: Drone Capable of Tracking Odour Developed
If the robot is commercialised, it is capable of revolutionised the Australian agricultural sector, helping to curtail the problem of labour shortage and low harvested apples.
The robot which uses highly sensitive vision system for identification can have the harvest time of the individual apple reduced to as low as seven second, the report says.
The led researchers, Dr. Chen stated that the in-house developed vision system would not just identify an apple within it reach but also able to class the obstacles within the vicinity of a fruit to calculate how it would manipulated it way to harvest the apple.
The machine vision system which works on deep learning is said use just 200 milliseconds to process image of an apple and is able to identify more than 90% of all apple seen within its camera view from a distance as far as 1.2m.
The vision system can operate in all type of lighting and weather weather conditions.
Share This