Meet Mahi Malhani: Class 12 Student who Built TRASHbot, an AI-Powered Waste-Sorting Robot

Inspired by litter scattered around dustbins during a school trip, Class 12 student Mahi Malhani developed TRASHbot, a robot that identifies and sorts waste with 90% accuracy. Now being tested in homes, schools, residential societies and public settings, her innovation demonstrates how young minds can use technology to address everyday civic challenges.

A school trip to Delhi’s Sundar Nursery in 2023 turned into the starting point of an innovative journey for Mahi Malhani. While exploring the heritage park, the then school student noticed a troubling sight: wrappers, plastic bottles and food waste scattered on the ground, often just a few steps away from dustbins.

The sight prompted her to think beyond the usual calls for cleanliness. What if technology could make waste disposal easier and encourage people to take greater responsibility for their surroundings?

“I noticed litter everywhere, even near dustbins. People often take the easy route and throw waste wherever they can. I wanted to bridge that gap with technology that helps, rather than just tells people to clean up,” Mahi recalls.

Today, the 17-year-old Class 12 student of Amity International School, Mayur Vihar, Delhi, has transformed that observation into TRASHbot, an AI-powered robot designed to collect and sort waste. With applications ranging from school cafeterias to residential communities, the project reflects how innovation can begin with a simple question about an everyday problem.

Turning a Civic Challenge into an Innovation

India’s mounting waste-management challenge provided the broader context for Mahi’s idea. According to a 2024 Press Information Bureau release citing the Central Pollution Control Board’s 2021–22 annual report, India generates an average of 1,70,338 tonnes of solid waste every day, of which 91,512 tonnes is treated.

For Mahi, the issue was not simply about the availability of dustbins. It was also about convenience, awareness and the gap between knowing what is right and actually doing it.

Her interest in technology had begun much earlier. Fascinated by gadgets and electronics, she enjoyed experimenting with devices, dismantling old equipment and exploring how machines worked. She began learning Python before moving on to C++ and JavaScript.

“I wanted to understand how machines think and move. If I can teach them to act intelligently, like humans, but more precisely, they can help solve everyday problems, like managing waste efficiently,” she explains.

In August and September 2023, she began researching possible solutions, sketching designs and exploring ways to automate waste collection and segregation. Her objective was to develop a system that would be practical, user-friendly and capable of operating with minimal human intervention.

From Sketches to TRASHbot

Moving from an idea to a functional prototype required technical guidance and experimentation. In August 2024, during a summer programme at IIT Delhi, Mahi came across Rancho Labs, a robotics and Internet of Things (IoT) initiative. Curious about its work, she approached the team and, with her parents’ support, participated in discussions and a survey to understand the practical aspects of robotics.

Mentors associated with Rancho Labs and her school computer teacher, Deepshikha Sethi, helped her understand technical challenges and identify possible improvements.

“They were always available to answer questions and suggest improvements,” Mahi says. “But the prototype itself was my own creation, and that independence was very motivating.”

The result was TRASHbot, a mobile robot designed not merely to act as a moving dustbin but to identify waste, navigate its surroundings and direct the waste into the appropriate compartment.

“I wanted it to feel alive, almost like a companion in cleanliness,” she says.

How the AI-Powered Robot Works

TRASHbot combines artificial intelligence, sensors, motors and two small computing systems to perform its tasks.

The robot can be controlled manually through a smartphone application or switched to automatic mode, allowing it to move independently. Ultrasonic sensors detect nearby obstacles and help the robot avoid collisions with walls, furniture and people.

For waste identification, TRASHbot uses a camera and an AI model called YOLOv3. The camera captures images of waste items, and the model analyses them to determine whether they belong to the categories the robot has been trained to recognise.

At present, the system separates waste into two broad categories: biodegradable materials, such as food scraps, and non-biodegradable materials, such as plastic wrappers.

Once an item is identified, motors operate the relevant compartment, allowing the waste to be directed into the appropriate section.

The robot’s electronic components divide the workload. A Raspberry Pi processes the camera images and supports the AI-based recognition system, while an Arduino Uno controls movement and coordinates the robot’s actions. Six motors help it move, while two additional motors operate the waste-sorting mechanism.

Despite this technical complexity, Mahi has designed the user experience to be straightforward. Users can guide the robot using the application or allow it to operate in automatic mode.

“It may sound complicated, but for the user, it is as simple as pressing a button or letting it work on its own,” she says.

From Home Trials to Residential Societies

The first trials took place at Mahi’s home before the robot was introduced to her school cafeteria, where it was tested during busy lunch breaks.

Dr SK Singhal, Head of the Physics Department at Amity International School, recalls the experience of watching TRASHbot operate among students.

“Seeing it move between tables and collect wrappers was incredible,” he says, noting that its sensors and LED lights attracted students’ attention and made the process of waste disposal more interactive.

The next stage involved testing the robot in residential communities. At Lotus Boulevard Society, joint secretary Gita Nath observed how the device could simplify waste segregation.

According to Nath, the robot automatically directs waste into the appropriate compartment, making disposal more convenient for residents.

Yash, a 17-year-old resident of the society, found it particularly useful during events involving food stalls, when waste could accumulate quickly. The device helped make responsible disposal easier for visitors.

Since March 2025, TRASHbot has reportedly been deployed in two residential societies in Noida and used in collaboration with a municipal corporation in Udaipur. Around 15 to 20 people have directly interacted with the device so far.

The trials have also helped Mahi understand how the robot performs outside controlled environments and how people respond to a technology-based approach to cleanliness.

Measuring Performance and Overcoming Challenges

TRASHbot has recorded encouraging results during its trials. According to the reported performance figures, it identifies waste correctly approximately 85% of the time, sorts waste into the appropriate compartment in 90% of cases and avoids obstacles with a success rate of 98%.

A fully charged battery provides approximately two and a half hours of operation, allowing the robot to complete multiple rounds before requiring recharging.

However, developing a functional robot was only the beginning. Making it reliable in everyday settings presented a series of technical challenges.

In the early versions, voltage fluctuations caused the Raspberry Pi to restart unexpectedly. Delays in Bluetooth communication also affected the time taken to recognise and sort waste. Uneven surfaces, rain and crowded environments created further challenges, requiring improvements to the robot’s mobility and sensors.

Cost was another important consideration. The initial prototype cost approximately Rs 20,000 to build. Following refinements, the cost was brought down to around Rs 7,000, with further reductions being explored.

“Every time the robot faced a problem, I had to find a solution that made it strong, safe, and efficient,” Mahi recalls.

These setbacks became opportunities to refine the design and improve its practical usefulness.

Recognition Beyond the Classroom

What began as a school project has also earned Mahi recognition in technology and innovation competitions.

In 2025, she participated in the World Robot Olympiad, securing first rank in Delhi and second rank in India in the virtual championship, according to the information provided about the project.

She was also a finalist at the Inter-Amity Science Fair, Vasudha. In addition, TRASHbot was selected among the top 50 AI-based projects from more than 3,500 entries nationwide at the India AI Impact Summit.

These experiences gave Mahi opportunities to present her work, respond to questions from judges and learn from other young innovators.

The project has also generated interest among residents and students. Its movement and LED indicators draw attention, while its interactive design encourages people to participate in waste disposal rather than treating cleanliness as someone else’s responsibility.

For Mahi, this behavioural aspect is as important as the robot’s technical performance. She hopes that making waste collection more accessible and engaging will help people develop better habits.

Expanding the Possibilities

Mahi’s plans for TRASHbot extend beyond its current capabilities. At present, the robot distinguishes between biodegradable and non-biodegradable waste. She hopes to expand its classification system to recognise five categories: plastic, metal, glass, biodegradable waste and non-biodegradable waste.

Other proposed improvements include Wi-Fi connectivity and cloud-based monitoring, which could allow schools and residential communities to track waste collection remotely. Solar panels may help extend battery life, while improvements to the application could make the robot suitable for larger public spaces.

Mahi is also considering how the device might operate in rural areas, where uneven terrain and limited infrastructure could require a different approach to mobility and deployment.

“The goal is to make TRASHbot effective everywhere, not just in cities,” she explains.

For now, the robot remains a developing solution, with further testing and refinement needed as its applications expand. Its reported performance offers an encouraging starting point, while its future effectiveness will depend on reliability, affordability, maintenance and suitability for different environments.

A Young Innovator's Answer to an Everyday Problem

Mahi’s journey began with a simple observation: waste often ends up on the ground even when dustbins are nearby. Instead of accepting this as an everyday inconvenience, she began exploring how technology could help bridge the gap between awareness and action.

TRASHbot represents the outcome of that curiosity, combining coding, robotics and artificial intelligence to address a civic challenge that affects communities across the country.

Her work also highlights the value of encouraging students to look beyond textbooks and apply their learning to real-world problems. With guidance from mentors, support from her family and the determination to improve her prototype through repeated trials, Mahi has taken an idea from sketches to practical testing.

The robot is not a complete solution to India's waste-management challenges, and responsible disposal will continue to depend on public participation and effective waste-processing systems. Yet it offers an interesting example of how young innovators can contribute to solving everyday problems.

For Mahi Malhani, the objective remains clear: make cleanliness easier, make waste segregation more accessible and use technology to encourage people to take greater responsibility for the spaces they share.

Sometimes, meaningful innovation begins not in a laboratory, but with a young person noticing something others have learned to overlook.