Smart Following Cart
Product design | 2024 | Berkeley, CA
Group Work
Original object and its intent:
We see self-driving cars on the roads and AI tools generating content in seconds.
Yet, people still carry heavy carts by hand—on streets, in parks, at airports.
That contrast made me wonder:If AI can drive, detect, and deliver—why are we still pulling carts by hand?
For Parents,
Traveling with a child means handling both the child and the luggage—at once.
For Outdoor Travelers,
Uneven roads and steep paths make dragging carts exhausting and risky.
What We Heard on the Ground
“I don’t usually pull the cart—my husband does that.But if I could help him, that’d be nice. Haha”
“On our trip, I didn’t just carry the luggage—I pulled my kid in the cart too.There was a slope… I honestly thought I wasn’t going to make it.”
“My back’s not what it used to be. Even grocery shopping with a cart is tough.But living alone… there’s no one else to do it.”
What’t the Real Problem?
It’s Not Just the Weight.
👵 Seniors lose balance and control when both hands are tied.
👨👩👧 Parents must open doors or manage kids—while pulling carts.
🧳 Travelers struggle to navigate crowds with one hand occupied.
What if it followed you, hands-free?
Solution
🛒 Follows the user autonomously
🧠 Understands context and adapts
🚧 Avoids obstacles in real time
🚪 Responds to human behavior (like door holding or hand gestures)
Aim For
🙌 Hands free—to hold a child, a door, or someone you love
💬 No need to control—the cart simply understands you
🌱 From mobility to presence—more ease, more connection
From Sensors to Smarts: How the Cart Thinks
What makes it “smart”?
• It detects more than just location. • It understands intent (pausing at doors, waiting for you). • It responds in real-time like a companion, not a tool.
Testing the Prototype: Can It Really Follow?
• ✅ 10 real-life test scenarios
• 🧠 8/10 correct intent recognition
• 🚧 ~90% obstacle detection success
• 🤖 85% smoothness rating (peer evaluation)
