I. Introduction
Safety of people in dangerous or inaccessible areas has gained
importance in industrial, environmental, as well as in disaster-management
practices. Chemical plants, underground tunnels, gas-leaking areas, waste-dump
locations, or the disaster-damaged areas can be very dangerous because of the
presence of poisonous gas, unsound constructions, or poor visibility.
The life putting conditions
are known to be associated with traditional manual inspection which poses a
life-threatening situation creating the necessity of automated or
remote-operated systems that can conduct environmental assessment even without
putting human lives at risk.
To overcome this difficulty, applications of mobile robotic
platforms with powerful sensing technologies have attracted a lot of attention.
By use of remote-controlled vehicles that are installed with gas sensors,
vision modules, and obstacle-avoidance systems, real-time data can be collected
and the vehicle can safely travel at a distance. In this project, an RC car is
improved with an MQ135 air-quality sensor, an MQ4 methane sensor, an IR
obstacle detection module, and a camera located on the car to test and assess
dangerous conditions.
These sensors together are capable of the detection of harmful
gas, the ability to go around obstacles as well as capture live visual data.
This system provides a safe, affordable, and efficient method of hazard
evaluation in locations that are unsafe or otherwise inaccessible to humans,
and is eventually useful in safer decision-making in life endangering
scenarios.
II. Literature Overview
MQ series Gas Sensing with Metal-Oxide Sensors. Metal-Oxide gas
sensors, including MQ135 and MQ4 are common in a low-cost environmental
monitoring due to their response to a variety of target gases and ease of
interfering with microcontrollers.
The MQ135 is usually
mentioned in the context of general air pollutants (ammonia, benzene, smoke, CO
2 proxies), and is often used in Arduino/IoT monitoring prototypes. MQ4 (as
well as Metal-Oxide) is applied to methane/CNG detection and leak-monitoring.
Calibration procedures (finding R₀ in clean air, using a load resistor,
converting Rs/R₀ to ppm) are described in manufacturer datasheets and a
variety of applied studies, but it is also noted that absolute accuracy is only
available when compared to reference instruments.
Current characterization studies and reviews have noted that low-cost
Metal-Oxide sensors can respond to relative changes and trends with accuracy,
however, that accuracy is influenced by the environmental factors (temperature,
humidity) of the sensor, the sensor-to-sensor variation, and cross sensitivity
to different gases.
Off-the-shelf Metal-Oxide sensors (including MQ4) that are
benchmarked as potential low-cost methane sensors are noted to have a future,
with temperature/humidity compensation suggested and regular recalibration of
the sensor recommended to be done in the field.
Sensing the obstacles in the mobile robots, the infrared (IR)
proximity sensors are also a low-cost and mature kind of short-range obstacle
detection sensors that are frequently utilized as additional sensors in small
mobile robots and tele-operated platforms. Comparisons of sensor suites
indicate that IR is effective at simple collision-avoidance, but with increased
range, sensitivity to ambient light, and reflections on surfaces,
better-performing options to make strong navigation include ultrasonic sensors,
LiDAR, and vision-based methods (SLAM) when the cost and complexity can afford
it.
The remote inspection and teleoperated hazard surveillance. There
are numerous instances of tele-operated and semi-autonomous platforms (to
explore hazardous areas) in the literature of robotics (industrial leak
detection, disaster response, nuclear/chemical site inspection). Infrastructure
inspection and environmental robotics surveys emphasize the importance of
multi-sensor (gas sensors, cameras, IMUs, GPS) integration, the importance of
dependable communications in the event of real-time teleoperation, and the
usefulness of teleoperation in combination with control towards safer and more
efficient inspections. More recent applied literature addresses the topic of
multi-robot and swarm systems to cover a wider area.
The recent reviews are in agreement that Metal-Oxide based mobile
systems with low cost are still suitable to trend detection and early warning,
but to use them to deploy regulatory grade concentration mapping careful
calibration, environmental compensation, and cross-validation with reference
instruments should be employed.
Reliability in communication, power control and navigation
(particularly in the open air or within the confined GPS denying environment)
is an ongoing area of engineering work. It is also widely suggested that adding
camera verification, time-stamped cloud logging, GPS/SLAM and sensor fusion
(temperature/humidity and Metal-Oxide and electrochemical sensors) could help
enhance the resilience and utility of hazard-monitors rovers.
III. Materials
and Methods
The
hazard-analysis rover is developed by means of the combination of various
hardware units that are chosen as robust, accurate, and able to work in the
environment under distant conditions. As a base platform, a typical four-wheel
RC car chassis offers both mobility and stability as well as sufficient
load-bearing capabilities. The sensors and the motors are connected to an
appropriate microcontroller (like Arduino/ESP32) that can then communicate
wirelessly. Data processing and transmission is also done by the controller.
The MQ135 Gas Sensor identifies various air contaminants such as ammonia,
benzene, CO2, NOx and smoke. It is employed to measure the quality of air and
detect some toxic air.
The MQ4 Methane
Sensor mainly employed in monitoring the levels of methane gas in industries, sewer
pipes, and leak contaminants. It is fast responding and dependable. The IR
sensor identifies the obstacles through radiation emission and reception of
infrared radiation. It helps the rover to avoid collisions when in motion.
A small camera
will offer real-time video transmission of dangerous regions. It increases the
awareness of the situation and aids in visual inspection. The car has DC motors
that are controlled by a motor driver module which allows it to move forward,
backwards, and turn. The basic workflow of the module is as Wireless
Communication Module which is remote operation and transmitting sensor data is
carried out with the help of Bluetooth/Wi-Fi module (depending on its design). The
system is powered by a lithium-ion or Li-Po battery that allows it to be used
significantly in the field.
The approach is
interested in integrating, communicating, and coordinating the work of sensors,
control systems and the platform of the rover. The microcontroller is
interconnected with all sensors, and it controls the input data and motor
functions. The camera module will be mounted to a reasonable height in order to
have a front view. The MQ135 and MQ4 sensors constantly monitor the quality of
the air and its methane level.
These analog values are transformed into digital ones using the
ADC on the microcontroller, and then it is compared in real time to calculate
the level of hazards. The IR obstacle sensor keeps an eye on the area and
transmits digital data in case an obstacle is perceived. On the basis of this
input, the microcontroller will be able to direct the motor driver to modify
movement either halting the movement or altering direction to prevent
collision.
The video is
transferred through streaming to the device of the operator in real-time. This
enables remote navigation and visual inspection of dangerous or restricted
spots, which is safe. The networks described below employ wireless control and
data transfer systems as the rover has remote control of its movement using
Bluetooth or Wi-Fi. At the same time, the values of the gas sensors and live
cameras are sent to the operator to analyse them. This is a two-channel
communication, which provides continuity in control and monitoring. When
deploying the rover, the rover is directed into the target environment. The
operator monitors are Real-time gas levels, Live video feed, Obstacle alerts.
The system allows
the sensors to detect dangerous gases early, spatial mapping of the dangerous
zones and explore hazardous places that are not safe to ordinary human beings. In
the controlled environment, the rover is subjected to accuracy of sensors,
communication reliability, mobility performance and response time. Cross
checking of sensor readings with reference values is done to ensure accuracy.
Fig 1.
Block Diagram
Fig 1 describes that the
Operator/Remote Controller manages the rover and it gets the live data using
wireless communication. The Wi-Fi/Bluetooth Module reads the operator commands
to the ESP32 and also relay the sensor readings. The MQ135 and MQ4 gas sensors
transmit signals about air-quality and the amount of methane to the ESP32
microcontroller to process. The IR Obstacle Sensor identifies the surrounding
objects and the Camera Module broadcasts a video to the operator through ESP32.
Its Motor Driver (L298N) is connected to the motors in hazard detector and
controlled by ESP32 instructions. The Power Supply supplies the required power
to the entire module that includes the sensors, ESP32, camera, and the motor
driver.
IV. Result and
Discussion
The
hazard-analysis rover was subjected to indoor controlled tests and semi-outdoor
tests to test its accuracy in sensing, its mobility and responsiveness. Both
MQ135 and MQ4 gas sensors showed evident changes in output voltage in response to
the variation in air quality. The MQ135 was more sensitive to pollutant gases
with voltage rising with the concentration of the gas level and the MQ4 sensor
reacted to the richness of methane.
Tests in obstacle
detectors verified that the IR sensor was very effective in detecting objects
within its perception and this would activate instant stoppage or diversion of
the rover. Encountering several obstacles on the way to movement was observed,
which proves that the sensor will avoid collisions and allow it to be used in
autonomous navigation. Continuous visual feedback was achieved by the use of
the live camera system which helped the operator to keep an eye on the rover
environment and confirmed the occurrence of gas detection by the visual
indicators of smoke or closed areas. In general, the results obtained overall
testify to the fact that the rover is able to combine gas monitoring, obstacle
detection, and visual surveillance in a single mobile platform.
The system was
effective in the identification of hazardous areas, detection of gas leakage as
well as navigation of limited or dangerous areas that should not be entered by
man. These results justify the ability of the rover to be a low-cost,
dependable tool of remote hazard survey and environmental monitoring.
Fig 2 shows the relationship between MQ4 sensor output and harmful gas
concentration. The plot exhibits a strong positive correlation, where
increasing sensor output from 1 to 6 units corresponds to a rise in gas
concentration from approximately 10 ppm to 400 ppm. This trend demonstrates the
MQ4 sensor’s sensitivity to varying levels of harmful gases.
Fig 2: MQ4
Response to Gas Levels
Fig 3 presents the
time-dependent voltage response of MQ135 and MQ4 gas sensors. Both sensors
exhibit an initial rise in output voltage, reaching peak values around 8–10
minutes, followed by a gradual decline as the gas dissipates. The MQ135 sensor
displays a higher peak voltage, indicating greater sensitivity compared to MQ4
under the same conditions.
Fig
3 : Gas Sensor Response Over Time
Fig 4 illustrates obstacle
detection count as a function of time over a 10-second interval. The detection
values fluctuate, indicating the intermittent presence of obstacles within the
sensor’s detection range. Peaks in the plot suggest moments of increased
obstacle proximity or multiple reflections detected by the system.
Fig
4 : Obstacle Detection Counts
Fig 5 compares the hazard detection output voltage of MQ135 and MQ4
sensors. The MQ4 sensor records a higher output voltage than MQ135, suggesting
a stronger detection response to the tested hazard condition. This comparison
highlights the relative sensitivity differences between the two sensors.
Fig 5: Hazard
Detection Comparison
The outcomes of
the hazard-analysis rover prove the success of the implementation of gas
sensors, obstacle-detection, and visual-monitoring in a small mobile platform
applicable to the hazardous and inaccessible conditions. The MQ135 and MQ4
sensors were reliable with smooth curves of responses that were associated with
the change of the gas concentration during the experiment. Their difference in
sensitivity shows that both the sensors are complementary to each other,
whereas MQ135 is more responsive to general air contaminants and toxic gases,
MQ4 is focused on detecting methane, which is required in industrial safety
settings. The plateau region of the sensor graphs also shows the stability,
which also signifies a steady performance of the sensor once a homogeneous concentration
of the gasses is reached.
The results of
the obstacle detection point to the significance of the real-time navigation
aid in the limited space or uncertain conditions. The immediate response and
quick detection is reduced, which made the IR sensor
minimise the risk of collisions, and thus the rover was able to navigate in an
otherwise cluttered space safely. This feature is imperative when the
deployment is much-needed in the disaster-prone regions, chemical stores,
underground tunnels, and areas with toxic gases leakage where visibility and
accessibility of people are restricted.
The live video
feed also enhanced situational awareness because the operator was able to
visually confirm environmental situation and match it to sensor data. This multi-channel
monitoring system will improve the efficiency in decision-making and minimize
the use of one type of sensor. The general functionality of the rover indicates
that it can be a useful early warning and inspection instrument in detecting
potentially dangerous areas without putting human beings in danger.
Although the
system operates successfully, it is possible to enhance it through the
improvement of autonomous path-planning, incorporation of GPS to deploy it
under an open environment, and additional sensors, including the temperature,
humidity, or CO sensors to make it a more complete hazard assessment. Such
improvements would make the systems more reliable, able to target a wider range
of applications, and semi-autonomous or fully autonomous navigation in
complicated environments.
VI. Conclusion
The
hazard-analysis rover developed using RC which is shown in this work provides
an effective and feasible solution to observe the environment that is
associated with high danger to human life. Though the combination of the MQ4
gas sensor and the MQ135 gas sensor, the system is effective in detecting
harmful gases and leaks of methane and can be used to check the safety of an
industrial environment as well as environmental monitoring. The IR obstacle
detection block provides safe navigation in the narrow or cluttered
environment, and onboard camera offers visual feedback in real-time, which is
constantly checked, which allows assessment of the situation remotely without
human participation.
The experimental
tests confirmed the capacity of the system to react correctly to the variations
in gas concentration, prevent obstacles, and remain stable in motion in various
terrains. This proves the stability, dependability and functionality of the
rover in risky conditions. The solution is the combination of sensory
information with live video, which improves the accuracy of decisions made by
the operators, making it possible to detect dangerous conditions early and
preventive measures to be taken in time.
On the whole, the
project presents a simple, mobile, and flexible platform that can accommodate
safety tasks in industries, disaster areas, underground, and contaminated
areas. Its modularity can be expanded upon in the future with more features
like GPS-based navigation, cloud data logs, complex AI-controlled obstacle
avoidance, and the addition of other environmental sensors (CO, temperature,
humidity). Currently, as these upgrades have been made, the system has great
possibilities to become an entirely autonomous hazard-monitoring robot that
would be applicable in large-scale and real-world applications.
References:
[1]
Han, J., Lu, F., & Yang, Z.
(2018). Gas sensing performance of MQ-series sensors: A review. Sensors and
Actuators B: Chemical, 257, 1031–1041.
[2]
Espressif Systems.
(2021). ESP32 Technical Reference Manual. Espressif
Systems.
[3]
Bhatt, P., & Shah, M. (2020).
Application of low-cost gas sensors for environmental monitoring. International
Journal of Sensors and Applications, 12(3), 45–52.
[4]
Patel, R., & Jadav, P. (2019).
Performance analysis of methane detection using MQ4 sensor. International
Journal of Advanced Research in Electrical, Electronics and Instrumentation
Engineering, 8(2), 120–126.
[5]
Verma, A., & Soni,
R. (2020). Design of IR-based obstacle detection system for mobile robots.
International Journal of Scientific Research in Engineering and Technology,
7(4), 150–154.
[6]
Singh, R., & Kumar, P. (2017).
Real-time hazard detection in unmanned ground vehicles. International Journal
of Robotics and Automation, 4(2), 55–62.
[7]
Kumar, V., & Rajesh, P. (2021).
IoT-based surveillance robot using ESP32. International Journal of Innovative
Technology and Exploring Engineering, 10(9), 230–234.
[8]
Rahman, M., & Noor, M. (2019).
Design and development of a wireless-controlled environmental monitoring robot.
Journal of Embedded Systems, 6(1), 22–29.
[9]
Pandey, S., & Gupta, A. (2022).
Real-time environmental monitoring using mobile sensor nodes. International
Journal of Electronics and Communication Engineering, 11(3), 89–95.
[10] Espressif Systems. (2022). ESP32-CAM: Video
Streaming and Face Recognition Guide.
[11] Sharma,
T. & Nikhil, P. (2020). Wi-Fi based live video transmission using
ESP32-CAM. International Journal of Computer Applications, 176(37), 10–15.
[12] Gupta,
P., & Yadav, S. (2020). Low-cost wireless robots for hazardous area
monitoring. IEEE International Conference on IoT and Applications, 112–118.