Conference PaperPDF Available

Developing an Intelligent Waste Sorting System with Robotic Arm : A Step towards Green Environment


Abstract and Figures

Waste management as well as sorting is a very crucial task to make the environment green and to ensure better (re)use of the resources. Bangladesh, because of its high density population, is facing enormous challenges to manage huge amount of wastes produced every day. So the purpose of this paper is to use the advancement of Information and Communication technology (ICT) to improve the waste management system and make lives better by providing a smarter way for waste sorting and management. In this paper, an intelligent system was proposed and developed for automatically sorting the waste to be used in context of Bangladesh. A light weighted experiment was carried out to evaluate the system performance. The experiment replicated with 11 objects (waste) of different size and types. The experimental results showed that the proposed system was reliable and achieved about 82% accuracy for the categorization of different kinds of waste.
Content may be subject to copyright.
International Conference on Innovation in Engineering and Technology (ICIET) 27-29 December, 2018
Developing an Intelligent Waste Sorting System
with Robotic Arm : A Step towards Green
Sadia Zahin Diya1, Rifat Ara Proma2, Muhammad Nazrul Islam, Tasmiah Tamzid Anannya,
Abdullah Al Mamun, Rizvi Arefeen, Saifullah Al Mamun, Ihtiaz Ishmam Rahman, Md Fazle Rabbi
Department of Computer Science and Engineering
Military Institute of Science and Technology
Mirpur Cantonment, Dhaka -1216, Bangladesh
Abstract—Waste management as well as sorting is a very
crucial task to make the environment green and to ensure
better (re)use of the resources. Bangladesh, because of its high
density population, is facing enormous challenges to manage huge
amount of wastes produced every day. So the purpose of this
paper is to use the advancement of Information and Communica-
tion technology (ICT) to improve the waste management system
and make lives better by providing a smarter way for waste
sorting and management. In this paper, an intelligent system was
proposed and developed for automatically sorting the waste to be
used in context of Bangladesh. A light weighted experiment was
carried out to evaluate the system performance. The experiment
replicated with 11 objects (waste) of different size and types.
The experimental results showed that the proposed system was
reliable and achieved about 82% accuracy for the categorization
of different kinds of waste.
Keywords—Waste sorting, sensor, robotic arm, automation,
green environment
Waste materials can be classified in different categories
based on their re-usable functionality [1], that includes, (a)
recyclable waste like paper, metal, plastic, glass, cardboard
are used to recycle and reuse; (b) organic waste are used to
make compost for agriculture; and c) non-disposable waste are
not reusable and need to be buried as these are harmful for
the environment. To ensure the better use of these resources,
to promote the cost-effective production and cultivation, and
finally to make the environment green, it is utmost important
to categories and manage the wastes in to different category.
Waste sorting contributes to recycling and saving energy.
The Aluminum Association [2] estimates that the energy saved
in recycling a single aluminum can, could be used to power a
television for 3 hours. Creating new plastic from raw materials
requires ten times more energy than the energy needed for
recycling plastic [3], [4]. Environmental Protection Agency
[5] estimates that producing a paper product from recycled
paper requires only 60 percent of the energy required to create
one from fresh wood pulp. Energy Administration Information
[6] reports that recycling a ton of paper can save 17 trees.
However, due to lack of any efficient way of sorting the waste,
most of the recyclable products are wasted. This turns out to
be very harmful for the environment in the long run.
However, in context of Bangladesh, a huge amount of waste
materials are generated every day due to its high population
density that have a population of approximately 163 million
people [7]. To manage these huge waste materials properly,
no effective and innovative means are used. Everyday a vast
portion of effort and resource of Dhaka City Corporation is
invested for properly collecting and processing waste. More-
over, sorting these wastes in to different kind of wastes are
crucial challenge to green the environment of Bangladesh; as
because, waste materials that are dumped in the landfill creates
bio-gas that mainly contains methane gas (CH4) which is
approximately 30 times more overpowering as a heat-trapping
gas than Carbon dioxide (CO2) [1], [8]. It is also harmful
for the water body that are around the landfill. Again, waste
generation in Bangladesh is increasing vigorously and it is
expected to reach 47, 064 tonnes per day by 2025. The total
waste collection rate in major cities of Bangladesh such as
Dhaka is only 37% [9]. Which is much less than what it
should be. Such poor management system can cause health
hazard and major damage to the environment. In Bangladesh,
thus a system of sorting waste materials to reuse or recycle
may help to save a lot of space, person-effort, air-pollution,
land-pollution, and money; which in turn may lead Bangladesh
to move a step forward towards making greener environment.
Therefore, the objective of this project is to develop an efficient
automatic system to sort out different kinds of waste to make
the environment greener.
The other sections of this paper are organized as follows.978-1-5386-5229-9/18/$31.00 © 2018 IEEE
The work focusing to waste sorting and management are
presented in section II. The conceptual design of the proposed
system is discussed in section III. Section IV presents how the
system was implemented followed by discussion on system
evaluation. Discussion and concluding remarks are presented
in section VI.
This section briefly summarizes the work related to waste
sorting and management. In 2016, Williams & Bentil [10]
introduced and implemented an automatic waste management
sorting unit using micro-controller. In this design they suc-
cessfully sorted organic and inorganic waste materials. To do
this they used gas sensor which sent data to micro-controller
and using this data garbage was differentiated and used in
a recycling plant later on. Another work was conducted by
Elfasakhany et al. [11] on an autonomous system capable of
sorting common recyclable materials, namely ferrous and non-
ferrous metals, plastic and glass into distinct waste containers.
A micro-controller combined the interfaces of all the system
components using programming to control all the system
Authors presented the results of a survey in their paper
which showed three alternative methods to sort wastes from
building construction site were presented in [12]. Construction
wastes typically consists materials like timber, plaster, iron,
cement, plastics etc. Thus, such system will not be useful
for detecting daily household wastes which includes a large
amount of organic materials. Lukka et al. [13] in 2014,
presented a recycler called ZenRobotics Recycler. This robot
could pick up wastes from construction and demolition (CND)
sites and put it on a conveyor belt. It could sort between wood,
stone and metal without any human sorter through machine
Paulraj et al. [14] proposed an algorithm to detect waste
material from an image (captured by thermal imaging camera).
A robot was also developed which was equipped with a
thermal imaging camera, a proximity sensor and a 5-DOF
robotic arm. In another work [15], visual features were used
as sorting criteria. In such case, optical sensor and laser beam
were used to detect materials based on their shapes, colors
and textures and mechanical methods to perform sorting.
A technology based on optical identification of fluorescence
signature was developed by Ahmad [16] in his work. The
system was capable of tracing and identifying different kinds
of plastic materials using three commercial tracers. However
sorting any other kind of material was out of its scope.
Chahine & Ghazal [17] designed and developed an auto-
mated waste sorting system with the use of an inductive prox-
imity sensor, a capacitive proximity sensor and a photelectic
sensor. Ang et al. [18] developed automated waste sorter in
2013, which had a mobile robot delivery system. The system
included a line follower robot which could pick up waste and
drop it on a trash can. The system was successful in collecting
the garbage with more than 80% accuracy.
Another system was developed by Russel et al. [19] which
could sort metal, paper, plastics and glass. For sorting metal
and glass conventional sensors were used and for sorting paper
and plastics a sensor using LASER and LDR was developed.
A weight sensor and counter was also there to find out the
amount of sorted materials.
In summary, though a number of system has been developed
for waste sorting or management but these are costly, required
high skill to operate and maintain, focused to specific kind
of waste, and developed mainly for developed countries. In
Bangladesh, no such technology is available to sort the huge
amount of waste cost-effectively, efficiently and effectively.
Even though studies have been conducted in Bangladesh con-
text, there is still no practical implementation. This research
thus aiming to developing an automatic waste management
system considering the economical and contextual situation of
A conceptual design to develop an intelligent automated
waste sorting system is discussed here. The proposed system
would have a number of features to overcome the limitations of
existing system used in Bangladesh. The features may include
the following:
1) Developing a robotic arm that will pick up garbage from
garbage bin and then put it on a conveyor belt. This part
of the system will let people not to touch garbage by
their hand.
2) The system will be capable to sort different types of
wastes in different baskets basing on their type.
3) A mechanism which will be used to differentiate among
different types of waste materials namely plastic,organic
and metal objects.
4) The system will also include a database that keeps track
of different types of materials that have been detected in
a period of time. This database will help to keep track of
how many items are segregated. This data can be useful
in recycling.
5) Apart from this, a mobile app will be developed to
operate the robotic arm easily and remotely.
A flow diagram to represent how the system may work is
showed in Figure 1. The system will start using a mobile app
that will trigger robotic arm to pick garbage from a container
and put it on the detection system to identify its type. The
sensors in the detection system will detect the object. Then
the objects will be classified between metal, plastic, organic
and others based on the sensors output. Depending on the type,
a rotating surface will rotate and bring the appropriate bucket
in front of the conveyor belt. After that, the object will be
dropped in its respective container at the end of the conveyor
The proposed conceptual design as discussed in earlier sec-
tion is materialized as depicted in Fig 2. The brief description
of developing each module are presented below:
Fig. 1. Flow chart of the proposed system
The system includes a robotic arm which is operated by
a mobile app. The structure of the robotic arm is made of
stainless steel. The joints of the arm functions with the help
of servo motor. The mobile app is basically an android app.
Java programming language was used to develop the app. This
allows to operate the servo motor of the arm, using a Bluetooth
module. The robotic arm is showed in Fig 3 and the user
interface of mobile app is showed in Fig 4. The mobile app
is created using MIT App Inventor and it supports android
version 2.1 and above.
The detection module includes three sensors- inductive,
voltage and IR (Infrared Ray) sensor to detect and differentiate
between different waste materials. The inductive sensor is ba-
sically a proximity sensor which detects only metallic objects.
It’s working principle is dependent on a coil and oscillator.
An electromagnetic field is created in the near surroundings
of the sensing surface by the coil and oscillator. A metallic
object in the operating area reduces the oscillation amplitude.
The rise or fall of such oscillation is identified by a threshold
Fig. 2. Block diagram of the proposed intelligent waste sorting system
Fig. 3. Robotic Arm
circuit that changes the output of the sensor. From the rise
or fall of the oscillation amplitude that is measures against a
threshold, the output of the sensor is determined. Thus with
the help of this sensor all the metallic objects are separated.
Differentiation between plastic and organic objects was done
by voltage sensor. Two aluminium foil strips were used to
provide voltage to the object as showed in Fig 5. The input
portion of inductive sensor was in between two foil strips so
that the object touches all of them together. As it is known that
Fig. 4. User interface of the mobile application
plastic is non- conductive element, no voltage will be passed
through it. That is why, output of the voltage sensor will be
zero. On the other hand organic objects are conductive. So
voltage will pass through it and output of the voltage sensor
will be a non zero value. The IR sensor is used to detect
the presence of any object on the detection system. When the
output of this sensor is high, the system starts detecting the
object type.
Fig. 5. Sensors used in the system
Conveyor belt and the garbage bins are the last module of
this system. Fig 6 shows the implemented conveyor belt. With
the help of two gears and a DC motor, rotation of the conveyor
belt was achieved in a downward motion. This allowed to
transfer the waste material in a bin depending on its type.
A servo motor was attached to a circular surface containing
different colored containers. Basing on the type of material, the
circular surface rotated and placed a specific colored container
Fig. 6. Conveyor belt
at the end of the conveyor belt allowing the waste to fall in it.
Along with this the total count of different types of elements
were saved in a text file. This file was written from the serial
output of the Arduino Uno. Whenever an item was detected
it was updated in the text file showing the latest amount of a
specific type of element.
The developed system was evaluated in an academic en-
vironment at Software Engineering laboratory of authors’
institute. To evaluate the effectiveness and efficiency of the
system, 11 types of items(waste) of different dimensions were
passed through the system for sorting into different types. The
outcomes are presented in Table 1. The results showed that 9
out of 11 items were detected accurately, i.e. the accuracy of
that system was 81.8%. The results also showed that when
given an object that was mixed with two or more elements,
the detection was not accurate. Also in case of voltage sensor,
sometimes for metal detection, it did not give correct results,
which was not observed with the inductive sensor as it gave
only 1 or 0 as output.
The output of the voltage sensor is showed in Fig 7. In
case of organic materials, the voltage sensor output fluctuates,
for reading 1, it was 3.39 volts and for reading 2 it was 2.01
volts, reading 3 it was 1.32 volts and so on. But for plastic
and metal, the output curve was almost constant around 5.00
and 0.00 volts respectively. However, preference was given to
the inductive sensor while detecting metal objects.
In this paper, an intelligent waste sorting system is designed
and developed to segregate three different types of materials.
Whenever an object is put on the detection unit the IR sensor
detects the presence and readings of inductive sensor and
voltage sensor are considered. This reduces the possibility
of unnecessary rotation of the servo motor which rotates the
bucket carrier disc. As sensors are sensitive, it is possible
for them to give a value even when there is no object to be
identified. The robotic arm moves in such a pattern so that
Item No Object Type Dimension(cm2)Voltage Sensor(Volts) Inductive Sensor IR Sensor Detected Type Success
1 Aluminium Foil Metal 6x6 4.91 1 1 Metal Yes
2 Watermelon Organic 6x1.5 3.39 0 1 Organic Yes
3 Bottle Cap Plastic 5x5 0.00 0 1 Plastic Yes
4 KeyRing Metal 3.5x3.5 5.01 1 1 Metal Yes
5 KeyRing Plastic 3.5x5.6 0.06 0 1 Plastic Yes
6 Badge(Metal and Plastic) Mixed 5.8x5.8 0.99 1 1 Metal No
7 Apple Organic 5x1.2 2.01 0 1 Organic Yes
8 Leaf Organic 6x3.2 1.32 0 1 Organic Yes
9 Coin Metal 2x2 5.09 1 1 Metal Yes
10 Cucumber Organic 2x1 0.99 0 1 Plastic No
11 Container Plastic 5x4 0.00 0 1 Plastic Yes
Fig. 7. Voltage sensor output
there is enough time to finish detecting the waste material and
also place the appropriate bucket in front of the conveyor belt.
All the time delays have been taken under consideration and
the system is designed accordingly.
Compared to other systems, this proposed system is made
up of less equipment and it is cost effective. Only three sensors
are used for the detection part - voltage sensor, IR sensor
and inductive sensor. None of these sensors are too expensive
which makes the detection process effective at a low cost. The
system also differs in the sense that it has a robotic arm to pick
up and drop the waste materials. This arm has the option to be
operated automatically and simultaneously it can be controlled
by a mobile app. None of the existing systems have mobile
app to control the functionality of the waste sorting system
The research has a few limitations as well and future work
may focus to alleviate this limitations. Firstly, a high fidelity
prototype of the proposed system was developed and tested
only in an academic environment, thus future work may pursue
to develop the concrete version of the system and assess its
performance in a real environment with a large amount of
waste. Secondly, we were not able to compare our system in
terms of power, voltage or any other related metrics practically
with any other existing system. This is something we wish to
do in the future. Thirdly, the proposed system can screen and
detect only one item at a time. So there lies scope for future
improvement in this regard as well. Finally, the system can
segregate three types of elements at present, thus feature to
detect more types of waste can be included to improve the
overall performance and efficiency.
[1] Why should we sort our waste? Available:, [Accessed: 03-
[2] Aluminum recycling, Available:
advocacy/topissues/aluminum-recycling, [Accessed: 31-Mar-2018].
[3] Mba polymers, Available:, [Accessed: 30-
[4] How much energy does recycling save, Available: -energy-recycling-save-79720.html,
[Accessed: 31-Mar-2018].
[5] Environmental factoids, Available:
/epawaste/conserve/smm/wastewise/ web/html/factoid.html, Mar.
2016, [Accessed: 30-Mar-2018].
[6] Paper recycling facts, Available:
recycle/paperrecycling- facts/, [Accessed: 8-Jun-2018].
[7] Bangladesh population, Available: /coun-
try/Bangladesh, [Accessed: 03-Apr-2018].
[8] P. University, a more potent greenhouse gas than carbon
dioxide, methane emissions will leap as earth warms, Available:,
Mar. 2014, [Accessed: 03-Apr-2018].
[9] Waste management in bangladesh, Available:;
Mar:2018; [Accessed : 03-Apr-2018].
[10] E. A. Williams and J. Bentil, Design and implementation of a
microcontroller-based automatic waste management sorting unit for a
recycling plant, American Journal of Engineering Research (AJER), vol.
5, no. 7, pp. 248252, 2016.
[11] A. Elfasakhany, A. Arrieta, D. Ramrez, and F. Rodrguez, Design and
development of an autonomous trash sorting system, GJ PA Sc and
Tech. 01i2, pp. 5664, 2001.
[12] C. Poon, T. Ann, and L. Ng, On-site sorting of construction and
demolition waste in hong kong, Resources, conservation and recycling,
vol. 32, no. 2, pp. 157172, 2001.
[13] T. J. Lukka, T. Tossavainen, J. V. Kujala, and T. Raiko, Zenrobotics
recyclerrobotic sorting using machine learning, in Proceedings of the
International Conference on Sensor-Based Sorting (SBS), 2014.
[14] S. G. Paulraj, S. Hait, and A. Thakur, Automated municipal solid
waste sorting for recycling using a mobile manipulator, in ASME 2016
International Design Engineering Technical Conferences and Comput-
ers and Information in Engineering Conference. American Society of
Mechanical Engineers, 2016, pp. V05AT07A045V05AT07A045.
[15] J. Huang, T. Pretz, and Z. Bian, Intelligent solid waste processing using
optical sensor based sorting technology, in Image and Signal Processing
(CISP), 2010 3rd International Congress on, vol. 4. IEEE, 2010, pp.
1657 1661.
[16] S. Ahmad, A new technology for automatic identification and sorting
of plastics for recycling, Environmental technology, vol. 25, no. 10, pp.
1143 1149, 2004.
[17] K. Chahine and B. Ghazal, Automatic sorting of solid wastes using
sensor fusion.
[18] F. Ang, M. K. A. R. Gabriel, J. Sy, J. J. O. Tan, and A. C. Abad,
Automated waste sorter with mobile robot delivery waste system, in De
La Salle University Research Congress, 2013, pp. 79.
[19] M. H. Russel, M. H. Chowdhury, M. S. N. Uddin, A. Newaz, and M.
M. M. Talukder, Development of automatic smart waste sorter machine.
... It mainly collects waste from the footpath. S. Z. Diya et al. [2] proposed a system that would automatically sort the waste using a robotic arm. The system monitors different types of elements through database management. ...
Full-text available
Because of the economic development of industries, factories produce huge quantities of waste daily, of which some are reusable and some are non-usable. So, waste collection and segregation become relatively important for such industries. However, Bangladesh is far behind other countries in this arena. Since improper management of waste leads to considerable loss of industries, waste management becomes a big challenge for an industry in Bangladesh. The main objective of this project is to automate this waste management process with minimum human involvement to reduce the cost and save time. For this purpose, an automatic robot is built on an IoT platform using appropriate sensors, control units, and servo motors controlled by Arduino Uno. This robot collects the waste even from a nearly unreachable place and also, sorts them into metallic and non-metallic waste. We believe that the proposed robot is capable of meeting the strict requirement of low-cost as well as better performing units made by the local industries and hence, boost the use of waste management service of our country.
... Several studies have been conducted in the field of robotics [6], leading to the development of robots designed to assist in drawing pictures. For instance, Nimara et al. [7] presented a novel draw bot implementation based on a Raspberry Pi development board. ...
... The EPA-SWMM model uses historical and projected rainfall events or design storm water events for simulation (Jiang et al. 2015), and it is also capable of hydraulic simulation, including drainage/channel/stream flow routing (Diya et al. 2018) (Fig. 9.2). ...
Strom water management under climatic vulnerability is becoming a significant concern in urban areas of the world. The analysis and design of an urban storm water drainage/channel/stream/nallah system under non-stationary extreme rainfall intensity is a challenging task in any area. Apart from this, more attention is required to develop possible methods and techniques for alternate design of storm drains in fast growing urbanization. This research describes a hydrological and hydraulic comparison between the Strom Water Management Model (SWMM) and Modified Rational Method (MRM) for design peak discharge and forecast peak flow estimation in a Mega Industrial Park (MIP) in Bidkin, Aurangabad, India. The Intergovernmental Panel on Climate Change (IPCC) CMIP5 models of Representative Concentration Pathways (RCP 2.6, RCP 4.5, RCP 6 and RCP 8.5) are used for the future climate change scenarios. Furthermore, the study is carried out over a 100 years return period in which period (1970–2015) is considered as a historical period and (2016–2057) and (2058–2100) as future periods. The results of this research indicate that the estimated highest annual peak discharge of the stream at catchment for historical time period and for two future time periods by the SWMM model are higher as compared with the MRM method. Overall results show that non-stationary extreme rainfall intensity and design peak discharge indicated an increasing scenario in the future periods, it will lead to challenges for simulation of peak flow and hydrologic design of urban drainage/channel/stream in the study area catchment.
... x T × 100% (4) where, ∈ { , , , }; ∈ ℝ ...
Conference Paper
Full-text available
The outbreak of the Covid-19 pandemic has resulted in a surge in the generation of medical waste. Due to the transmissible nature of the Virus and the lack of effort at proper disposal, the safety of the front-line health workers, as well as the disposer, is at risk. Hence, to mitigate the spread of infectious diseases, a system is proposed that uses a robotic arm for segregating medical waste automatically. The robotic arm is operable through voice commands, and the segregating operation could function in automatic and manual mode. The system uses the YOLOv3 (You Only Look Once) algorithm to detect and classify the medical waste and then uses the Robot Operating System (ROS) platform to pick up and drop the waste object into color-coded bins. For this research, the medical waste has been categorized into 4 types, and for each type, a color-coded bin has been used for segregation. Our system has achieved 94% training accuracy for the YOLOv3 model on a custom dataset, whereas the system's overall accuracy in automated mode was 82.1%, derived after 30 trials. In the case of manual mode, an average accuracy of 82.5% has been achieved for the same number of trials.
Full-text available
Waste management is a massive issue in India, most of the present systems cannot manage waste on a scalable level, thus creating pressure on the ecosystem. Before the elimination of waste, segregation needs to be done to manage individual types of waste. Hence taken the same approach to solving the problem, which most of the present-day systems fail to do. The goal is to segregate the garbage generated in individual households into solid, liquid, biodegradable, non-biodegradable, combustible, and non-combustible, using many subsystems that involve electro pneumatics, compression, and storage. Image processing techniques will further advocate the process. The desired system will further reduce the waste of an in-built pulverizer. After conducting in-depth research on the present solutions for the urban waste processing chain, the level of complexity increases as the waste goes further along the chain and, in the end, the only option left is incineration was figured out. The solution allows endpoints of the chain to process different types of garbage in a more organized fashion. Municipal solid waste (MSW) is solid waste that results from municipal community, commercial, institutional, and recreational activities. This paper aims to segregate the MSW generated by households into biodegradable, non-biodegradable, combustible, and noncombustible. © 2023 Institute of Advanced Engineering and Science. All rights reserved.
Full-text available
An essential component of a waste management system is waste sorting. Correct sorting of waste is crucial for creating a clean environment for everyone, reducing pollution and reusing recyclable materials. Manual waste sorting can cause serious health issues to the workers due to various disease-causing agents present in the garbage. The use of robots in sorting of materials such as glass, paper, plastic, metals, etc., from other waste can facilitate the production of secondary raw materials as well as conserve energy and production costs. Robots can help in efficient sorting of waste and can work endlessly thus eliminating health hazard to labours. By using computer vision, artificial intelligence, and automation systems, both, the efficiency and accuracy of waste sorting can be increased. This paper showcases the recent studies related with implementation of robots in waste sorting for efficient recycling, algorithms, types of grippers and its advancements. Various studies have been discussed related to computer vision and comparison of various algorithms is also presented. Major challenges faced in implementation of robotic sorting on global scale and its future scope has also been discussed.
Full-text available
The article provides a brief overview of the most popular and promising methods of sorting municipal solid waste, attention is paid to automated systems for recognizing garbage objects during sorting, and in particular, their high cost with relatively low efficiency. The problem of using high-cost systems for recognizing waste objects by the payback factor with a low content of useful components in the total mass of garbage has been identified. A method of an automated garbage sorting system is proposed and described, in which a person is used as a system for recognizing garbage objects. This method significantly reduces the cost of electron-optical and tactile detectors that recognize garbage objects, while not requiring a huge database to compare 3-d images. At the same time, the recognition accuracy increases significantly, and the cost of the process of recognizing garbage objects on the conveyor is significantly reduced, which in general ensures a quick payback of the garbage sorting system. It is proposed that such a system be used as an intermediate stage in the improvement of image recognition systems for sorting garbage objects.. The approach of using the human ability to recognize as a more economical method to recognize waste objects when sorting them in robotic systems is proposed. At the same time, comfortable working conditions are provided for a person, a sorter operator, who recognizes and encodes each sorting object remotely from garbage using video cameras and a high-resolution monitor. A variant of such a sorting system using mechanotronics is described.
The creation and management of an intelligent waste sorting system using a robot arm are described in the study. This system comprises a 6DOF robot manipulator, a machine vision block, and a control unit for rubbish sorting using analytic pictures. The item will be recognized by the neural network using the YOLOv4 software. This technique detects and identifies various types and kinds of trash, including paper-based waste, metal waste, and plastic waste. Results of offline testing on a library of more than 600 untrained photos reveal that the trained model has an average classification accuracy of roughly 98.43%.KeywordsWaste sortingRobotic armYOLOMachine vision6DOF
Full-text available
Nowadays solid waste is generated at an unprecedented rate due to rapid urbanization and industrialization. In the developing countries, recycling of useful materials from solid waste such as wood, plastic, glass and metal is severely constrained by limited door-to-door collection and poor means of waste sorting. Indeed, recovery of useful materials still relies on manual sorting, which is highly risky and hazardous for health. This paper details the design and implementation of an automated solid waste sorting system. The developed system is equipped with a programmable logic controller, an inductive proximity sensor, a capacitive proximity sensor, and a photoelectric sensor. Multi-sensor data fusion is used so that each material results in a different combination of sensor outputs. Once a material is identified, the controller makes a stepper motor rotate a defined angle to collect the material in its bin. Testing results show that the system is successful in sorting the four different materials.
Conference Paper
Full-text available
Municipal solid waste (MSW), generated at an unprecedented rate due to rapid urbanization and industrialization contains useful recyclable materials like metals, plastic, wood, etc. Recycling of useful materials from MSW in the developing countries is severely constrained by limited door-to-door collection and poor means of segregation. Recovery of recyclables is usually performed by waste pickers, which is highly risky and hazardous for their health. This paper reports the development of a robotic mobile manipulation system for automated sorting of useful recyclables from MSW. The developed robot is equipped with a thermal imaging camera, proximity sensor and a 5-DOF robotic arm. This paper presents an approach for sorting based on automated identification from thermographic images. The developed algorithm extracts keypoint features from the thermographic image and feeds into clustering model to map them into a bag-of-word vectors. Finally, Support Vector Machine (SVM) classifier is used for identifying the recyclable material. We used the developed algorithm to detect three categories of recyclables namely, aluminum can, plastic bottle and tetra pack from given thermographic images. We obtained classification rate of 94.3% in the tests. In future, we plan to extend the developed approach for classifying a wider range of recyclable objects as well as to incorporate motion planning algorithms to handle cluttered environments.
Conference Paper
Full-text available
Modern world meets lots of challenges that includes Smart waste management system. It is become matter of big concern if proper disposal system is not managed. Managing waste effectively and recycling efficiently, a nation can ahead one step forward. In this work, an automatic sorter machine is developed which can sort out the wastes in various categories to make waste management easier and efficient. It can be possible to sort out metal, paper, plastics and glass by developing an electromechanical system using microcontroller and operational amplifier. For sorting metal and glass conventional sensors are used and for sorting paper and plastics a sensor using LASER and LDR is developed. A weight sensor and counter is used to find out the amount of sorted materials. By using the proper recycling system, the curse of waste will turn into blessings for the civilization. The sorting procedure will make recycling more efficient. By means of this waste sorter, the conventional waste management system will be transformed into SMART system. This SMART system will help to make our environment more suitable for living, reducing global warming and making the world healthier.
Full-text available
The aim of this work is to design, build and test an autonomous system capable of sorting common recyclable materials, namely ferrous and nonferrous metals, plastics and glass into distinct waste containers. The autonomous system uses a monitoring system based on advanced sensor and classification techniques, which can improve the accuracy and reliability of sorting procedures. The system was designed with an aluminum chassis, loading carrier moved by a geared motor and a rubber band, sensors, control board, air fan for blowing the plastic bottles, DC motor to push the sorted material, and containers for the sorted materials. Electromagnetic sensors used electrical and magnetic properties to distinguish different metals. An air fan and a light sensor are used to distinguish between plastics and glass materials. A microcontroller combined the interfaces of all the system components using programming to control all the system actions. Diverse iterations were used for distinguishing different trash materials. The advantages of this system are that it is simple, light, efficient, environmentally green and inexpensive. The system is able to sort trash materials of different shape, size, weight and colour without being effected by dust, coatings or other impurities covering the trash materials.
Solid wastes are always collected as mixtures of different materials. They gets crushed, classified and sorted in solid waste treatment plants. Among these processes the sorting is the determining step for recycling and reuse. Traditional sorting technologies like magnetic sorting and eddy current sorting are only able to process some special kinds of ingredients of waste mixture roughly, such as the separation of ferrous and non-ferrous metals. Since there exist corresponding force fields between waste particles and separators. Some other properties of the solid particles such as the colours, shapes and texture features could also be considered as sorting criterions but there is no sufficient force field between these properties and separators. In this paper, an indirect sorting process by using optical sensor and mechanical separating system was developed and introduced. By using this system the particle sizes and positions, colours and shapes of each waste particle are able to be determined and used as sorting criterion. The mechanical sorting device consists of a compressed air nozzle which is controlled by computer, the target particles which were recognized by sensor were blown out of the main waste stream. Feature recognition by using optical sensor yield good results. This research provides a new approach for multi-feature recognition of sensor based sorting technology.
The construction industry is the major solid waste generator in Hong Kong. In 1998, it generated about 32 710 t per day of construction and demolition (C&D) waste. In the management of such a huge quantity of C&D waste, Hong Kong has adopted a strategy of depositing the inert portion (e.g. sand, bricks and concrete) of the waste at public filling areas for land reclamation and the non-inert portion (e.g. plastics, paper, wood) at municipal solid waste landfills. However, the C&D waste arisen is usually in the form of a mixture of both inert and non-inert materials. As a result, the waste has to be disposed of at landfills, aggravating the landfill shortage problem. There is a paramount need to separate the C&D waste into its constituent parts before it is delivered to either the landfills or the public filling areas for disposal. In order to study the feasibility of carrying out on-site waste sorting and the current situation of the building relating C&D waste generated in Hong Kong, a survey was conducted. This paper presents the results of the survey undertaken to evaluate three alternative waste sorting methods on building construction sites and to compare them with the use of an off-site central waste sorting facility. The results indicate source separation has the advantages of requiring less effort and resulting in better segregation of inert and non-inert wastes as compared with waste sorting centrally carried out at a designated area on- or off-site. In addition, the views of the building industry participants were also obtained through a questionnaire survey to give a better understanding of their attitude on on-site waste sorting. The results indicate that the building construction participants are reluctant to carry out on-site waste sorting. Even when high a tipping fee is imposed, they have little incentive to perform on-site waste sorting which is considered to be time and labour demanding. Only through contractual requirements or legislation can on-site waste sorting be fully implemented and becomes a long-term solution to the landfill shortage problem in Hong Kong.
A new technology for automatic sorting of plastics, based upon optical identification of fluorescence signatures of dyes, incorporated in such materials in trace concentrations prior to product manufacturing, is described. Three commercial tracers were selected primarily on the basis of their good absorbency in the 310-370 nm spectral band and their identifiable narrow-band fluorescence signatures in the visible band of the spectrum when present in binary combinations. This absorption band was selected because of the availability of strong emission lines in this band from a commercial Hg-arc lamp and high fluorescence quantum yields of the tracers at this excitation wavelength band. The plastics chosen for tracing and identification are HDPE, LDPE, PP, EVA, PVC and PET and the tracers were compatible and chemically non-reactive with the host matrices and did not affect the transparency of the plastics. The design of a monochromatic and collimated excitation source, the sensor system are described and their performances in identifying and sorting plastics doped with tracers at a few parts per million concentration levels are evaluated. In an industrial sorting system, the sensor was able to sort 300 mm long plastic bottles at a conveyor belt speed of 3.5 m.sec(-1) with a sorting purity of -95%. The limitation was imposed due to mechanical singulation irregularities at high speed and the limited processing speed of the computer used.
a more potent greenhouse gas than carbon dioxide, methane emissions will leap as earth warms
  • P University
P. University, a more potent greenhouse gas than carbon dioxide, methane emissions will leap as earth warms, Available:, Mar. 2014, [Accessed: 03-Apr-2018].
Design and implementation of a microcontroller-based automatic waste management sorting unit for a recycling plant
  • williams
E. A. Williams and J. Bentil, Design and implementation of a microcontroller-based automatic waste management sorting unit for a recycling plant, American Journal of Engineering Research (AJER), vol. 5, no. 7, pp. 248252, 2016.
Design and development of an autonomous trash sorting system
  • A Elfasakhany
  • A Arrieta
  • D Ramrez
  • F Rodrguez
A. Elfasakhany, A. Arrieta, D. Ramrez, and F. Rodrguez, Design and development of an autonomous trash sorting system, GJ PA Sc and Tech. 01i2, pp. 5664, 2001.