Inspiration
Something as simple as dropping your keys, phone, or wallet can become a painful or even dangerous task for someone with back problems, hip injuries, or limited mobility. For many of us, bending down takes only a second. For others, it can mean discomfort, loss of balance, or needing someone else's assistance. We wanted to use robotics to solve a problem that often gets overlooked because of how ordinary it seems. That's how AutoPicker was born. A robotic assistant designed to make everyday life more accessible, one dropped item at a time.
What it does
AutoPicker is an assistive robotic rover designed to locate, retrieve, and raise dropped objects to a comfortable height for the user. Instead of requiring someone to bend down, AutoPicker brings the object back up to them. Our goal is to make the entire process as effortless as possible, especially for elderly individuals, people recovering from injuries, and those experiencing back or hip pain. Beyond convenience, AutoPicker is about helping people maintain their independence when performing everyday tasks.
How we built it
AutoPicker combines robotics, computer vision, and mechanical/electrical engineering to tackle three main challenges: finding an object, retrieving it, and lifting it. We designed the system around a mobile rover that can navigate toward a dropped item. Its vision system is intended to identify and locate objects, while a retrieval mechanism allows the rover to pick them up. A lifting mechanism then raises the retrieved object to a more accessible height. The project brings together software and hardware, requiring us to consider object detection, movement, positioning, mechanical design, and how the different components communicate. Rather than building a robot simply because we could, we focused on designing one around a practical human need.
Challenges we ran into
One of our biggest challenges was connecting the software and physical components into a system that could reliably interact with real-world objects. Recognizing an object is one thing, but determining its position, approaching it correctly, and physically picking it up introduces an entirely different level of complexity. Another important challenge was balancing functionality with accessibility. The robot needed to do more than retrieve an object; it needed to return that object at a height that would genuinely reduce physical strain. These challenges pushed us to think beyond individual features and consider how the entire system would work together.
Accomplishments that we're proud of
We're proud of approaching robotics from an accessibility-first perspective. AutoPicker isn't about making a flashy robot. It's about exploring how robotics can give people more independence in their daily lives. We're also proud of bringing together multiple engineering disciplines around one shared objective. The project challenged us to think about computer vision, robotics, physical interaction, and user experience as interconnected parts of a single solution. Most importantly, we developed an idea with the potential to make a meaningful difference for people who face physical challenges that others rarely think about.
What we learned
AutoPicker taught us that building a robot capable of interacting with the physical world is significantly more complicated than developing software alone. We learned to think about the relationship between perception and action. A robot must not only understand where an object is, but also determine how to reach it and manipulate it safely. We also gained a deeper appreciation for accessibility-focused engineering. Sometimes the most meaningful innovations don't come from solving enormous, futuristic problems. They come from recognizing small everyday difficulties and finding ways to make life easier.
What's next for AutoPicker
Our long-term vision is to turn AutoPicker into a reliable, affordable assistive companion that can operate in everyday environments. We want to improve object recognition, expand the variety of objects it can retrieve, and make its navigation and lifting mechanisms more precise. Eventually, we envision AutoPicker operating autonomously in for example nursing homes, responding when someone drops an item and returning it without requiring physical effort. We also hope to explore additional assistive features that could make the rover useful beyond object retrieval.
Our ultimate goal is simple: make independence more accessible, one pickup at a time.
Log in or sign up for Devpost to join the conversation.