FARMING is embracing a technological revolution that involves the likes of computers, global positioning systems (GPS), unmanned drones, driverless tractors, and robots and much more.

The industry is on the brink of enjoying major developments in robots that could take the hard work and drudgery out of harvesting fruit and vegetables by hand.

There are two critical factors that separate a robot from a machine - the capacity to be autonomous or to act with some degree of decision, and presence of actuators, which allow a machine to alter its environments. This precludes computers and fridges from being called robots.

Researchers are using their knowledge of soft robotics to create deformable manipulators, or "hands", that can manipulate "fragile" produce, such as broccoli, strawberries and mushrooms. Currently, UK researchers are working on a harvesting robot that can handle and cut iceberg lettuces with the same handling care as human harvesters.

The technology has the potential to be deployed at any time of day and impact positively on the productivity and life quality of agricultural workers.

Researchers now envisage robots that could perform multiple tasks. For example, inter-changeable tools would allow switching between tasks such as seeding, tillage, spraying and harvesting.

You could also have robots for agriculture and food production that would perform other useful tasks at the same time such as surveillance, keeping a watchful eye on crops, livestock and expensive farm machinery, while carrying out their primary duties on the farm or in the factory.

Satellite imaging is used a lot in agriculture nowadays. A drawback with that technology is that the images aren't as good when there is cloud cover, but low-flying drones can consistently produce high resolution pictures of crops and their condition.

Different sensors - visible light, infra-red and thermal - are being used to identify different features of crops.Researchers have created a database that, for example, recognise weed types, the presence of disease, plant stress levels, crop damage, and crop yield potential.

Identifying weeds, their density and precise location within a field could, with a GPS-programmed sprayer, result in localised spraying and a big saving in chemical costs. Similarly, by identifying chlorophyll content in the crop plants the technique could be used for variable rate nitrogen fertiliser application.

"EyeSpot", a research project part-funded by AHDB Horticulture, is driving innovation in weed control and aims to reduce herbicide inputs for field vegetables by up to 95 per cent.

A team of experts at University of Reading are developing the use of a cutting-edge automated spot herbicide ejector, which will "point and shoot" metered droplets to individual leaves of unwanted plants in row crops.

The ejector will use an innovative imaging system to distinguish weeds in field vegetable crops and will evaluate the dose of droplets required to kill weeds at different growth stages. Herbicide droplets will then be accurately targeted to the leaves of the unwanted plants. The concept is "no herbicide applied to the crop; none to the soil; only to the weeds".

The project has been developed in response to concerns about the loss of herbicides through legislation and pressure to target pesticides better and in lower doses. The technology will precisely apply herbicide only to the weeds in such a way as to eliminate drift and splatter, while minimising the likelihood of run-off to soil and non-target organisms, including the crop. By applying droplets of a systemic, non-selective herbicide to individual leaves, it represents a paradigm shift in weed technology, achieving selective weed control by engineering rather than chemistry.

An automated vision-guided droplet application system should be ready for preliminary trials in 2018.

Livestock farmers are also using advanced technologies and robots. More and more dairy cows are being milked by stationary "robots". After some initial training, cows learn to stand over the robotic milking machine. They are encouraged to do so by the prospect of having their udders relieved of milk as well as a measured portion of tasty, concentrated feed being offered to them from an automated dispenser.

Once in place, the robot uses electronic sensors to locate the cow's teats, attach the cluster and then milk the cow. Meanwhile, a computer recognises the cow by means of an electronic transponder fixed to her neck, records the amount of milk she gives and dispenses an amount of feed appropriate to her yield. It also knows not to feed a greedy cow that has come forward too often.