11
speeddataprocessingandtransmissioncapabilitiesofferedby irrigation systems have been developed, such as sprinkler irri-
5G or even 6G wireless connections. gation [256], [257], drip irrigation [258], [259] and capillary
The interconnectivity solutions address key biotic and abi- irrigation [260]. Kumar et al. [261] developed an IoT-based
otic factors influencing farming systems, offering solutions drip irrigation system comprising of capacitive soil moisture
across various applications. This entails integrating thermal sensor, DHT11 ambient temperature and humidity sensor,
camerasystemsandsoilmoisturesensorstodetectwaterstress DS18B20 soil temperature sensor, ESP-32 microcontroller,
for informed irrigation management [249]. Automated insect ESP8266Wi-Fimodule,pump,solenoidvalveandsolarpanel.
trapcamerasystemswillbeemployedtomonitorpestpressure Similarly, Jenitha et al. developed a Deep Bi-directional Long
and guide control strategies. Multispectral sensors will aid Short-TermMemory(DBLS-TM)algorithmtopredictthesoil
in identifying nutrient deficiencies and detecting weeds. All moisture and rainfall status according to the air temperature,
these elements can be linked to a cutting-edge 5G or 6G humidity, soil moisture, rainfall status, and wind speed col-
wireless system. Furthermore, multiple sensor platforms can lected by IoT sensors [262]. The volume of irrigation can
incorporate sensors onto fixed mounts, as well as multi-robot be calculated by predicted soil moisture, rainfall status and
ground and aerial platforms [250]. evapotranspiration level.
The adoption of multi-robot collaborative Simultaneous 2) Fertilizers and Pesticides management: Fertilizers can
Localization and Mapping (SLAM) [251][252] technology provide all nutrient requirements of plants, crops, and soil
holdsimmensepromise,particularlyinagriculturalproduction fertility [263]. Farmers can precisely know the exact amount
and environmental monitoring. In agriculture, diverse robots of nutrients required for their crops to save fertilizers and
can be equipped with distinct sensors to perform specific minimizeharmtotheecosystemcausedbyexcessfertilization
monitoring tasks, enhancing overall efficiency. For effective usingIoT-basedfertilizationsystem.Swaminathanetal.[264]
collaboration, real-time communication among robots (both collected the data from optical sensors, weather measuring
groundandaerial)isessential,necessitatingswiftinterconnec- sensorsandnitrogenphosphorusandpotassium(NPK)sensors
tivity. For instance, aerial robots can capture imagery to build to predict fertilizer recommendation. The trained Bi-LSTM
crop field maps and plan paths for ground robots, boosting prediction model performs effectively and produces better
efficiency in large fields. This entails efficient communication outcomes, which is close to the expert’s advice. At the same
infrastructure to cater to multiple users, including robots and time, pesticides are also important during crop management,
humanoperators.Roboticarmsmountedonaerialandground which can reduce the impact of weeds and pests and improve
robots can be used for specialized actions such as harvesting. crop productivity . To achieve precision weed spraying, Mary
In essence, IoT in agriculture is to establish robust, high- et al. [265] developed an IoT-based weeding robot equipped
speed, and responsive interconnectivity systems, fostering with an ESP32 AI camera to capture the photo in the field,
multi-sensor and multi-platform communication across farms. a NodeMCU to detect the weed and a servo motor to control
Thisconnectedsmartfarmingconceptencompassesdatafrom the pump of the nozzle for precision spraying. TFL Classify,
various sensors, algorithms translating data into actionable a Real time image classification powered by TensorFlow Lite,
decisions,androboticplatformsimplementingthesedecisions was adopted to detect weeds. For pest control, Azfar et al.
for profitable, sustainable, and efficient farm management. [266]setinfrared(IR)lightwallaroundthecottonplant.Once
This holistic approach is underpinned by an interconnected the insect obstructed the light, the position-sensitive detector
IoT system that facilitates data transmission, processing, in- can catch the light deviation. The detection coordinates were
telligence, and the management of robotic fleets. sent to the drone to respond by spraying pesticide in the
Smart agriculture is a management concept implemented detection region.
with advanced technology, such as big data, the cloud com- 3) Microclimate management: Greenhouse production is
puting,artificialintelligence,roboticsandtheinternetofthings consideredasanultimatesolutionforincreasingfooddemands
(IoT) for monitoring, tracking, automating, and analyzing spurred by the growing population. Greenhouse offers a year-
agricultural operations [253]. The IoT is one of the important round production environment for fresh vegetables, boasting
technologiesinsmartagriculturethatcancollectdataandcon- a production rate approximately 50% higher than open-air
nect devices. It can provide a diverse set of tools for farmers cultivation [267]. Compared to other agricultural industries,
to address several challenges in the field [254]. Farmers can the commercial greenhouse consumed largest energy [268].
remotely access and manage their farms from anywhere at Therefore, it is important to manage the microclimate pre-
any time using IoT technologies. The utilization of cameras cisely in the greenhouse to save energy and cost. Ullah et
and sensors first collects valuable data and then uploads it to al. [267] collected temperature, humidity, CO2 concentration,
the cloud. After data analyzing on the cloud, actuators are solar radiation and wind speed using sensors to monitor the
used to regulate farming processes automatically. Farmers can microclimate in the greenhouse. The developed optimization
achieve these data using smart phone or PC to monitor the scheme can be used to control seven greenhouse actuators
farm. Many IoT applications in smart agriculture have been (heater, chiller, dehumidifier, fogging system, CO2 generator
studied in literature. and forced and natural ventilation) and achieve a tradeoff
1) Irrigation management: The irrigation system collects between energy consumption and plant growth rate.
the data from soil, climate and plant using IoT technology 4) Plant stress detection: Detection of plant stress at early
to calculate the water requirements of crops and adjust the stage is important for improving the crop production because
water flow to avoid water waste [255]. Several IoT-based plant stress can lead to crop diseases and death [269]. The