GRACE
Geriatric Robotic Assistance for Care and Engagement is a mobile robot focused on intelligent monitoring, daily reminders, and dedicated care for elderly users.
No arms. No lifting grandma. Just brains and vibes. 🧠✨
Meet GRACE
A friendly companion designed specifically for elderly care. GRACE doesn't have arms or lift anything; instead, she relies on advanced perception and AI to keep seniors safe, engaged, and independent at home.
The Unprecedented Care Crisis
With the global senior population projected to double by 2050, how can we develop a scalable, non-intrusive solution?
Companionship-First Robotics
Built for Elderly Care
GRACE autonomously navigates indoor spaces, follows humans, monitors health vitals via a smart wristband, detects posture anomalies, and provides verbal prompts for exercises, medication reminders, and daily routines.
She instantly sends caretaker alerts when anomalies or emergencies are detected, bridging the gap between elderly independence and continuous remote monitoring.
From Sketch to Reality
GRACE's design evolved from pencil sketches to 3D renders to a physical prototype — every iteration refined for elderly interaction.
Advanced Visual Perception
A multi-sensor perception stack that maps, scans, follows, and understands the environment in real-time.
RPLidar A2M7
360° laser scanning for SLAM-based navigation and real-time obstacle avoidance. Maps the entire environment.
Intel RealSense D435i
RGB-D stereo camera with depth perception. Used for human detection, following, and spatial awareness.
Standard RGB Camera
General-purpose vision for pose detection, context awareness, and AI inference running on the Jetson GPU.
Ultrasonic Sensor
Close-range obstacle detection as a safety fallback. Works even in conditions where LiDAR may miss objects.
What Can GRACE Do?
SLAM Navigation
Autonomous mapping and navigation using SLAM Toolbox and Nav2. GRACE builds, saves, and navigates indoor maps independently.
Human Following
Detects and follows a person using depth + vision, maintaining a safe distance while moving through the environment.
Posture Detection
AI-powered pose estimation to detect falls, slouching, or unusual positions and alert caretakers immediately.
Health Monitoring
Integrates with a smart wristband to track SpO2, heart rate, and lifestyle metrics in real-time.
Verbal Interaction
Daily reminders, exercise suggestions, and medication prompts for the elderly.
Caretaker Alerts
When anomalies such as abnormal vitals, falls, or distress are detected, GRACE instantly notifies the remote caretaker.
A Walking Nursing Station
GRACE collects and processes four categories of real-time data streams, turning raw sensor data into actionable care insights.
Robot Health
Hoverboard 40V & 24V battery systems, discharge and charging currents, ESP32 board temperature, and core telemetry.
Person Vitals
Smart wristband integration tracking heart rate, SpO2, systolic/diastolic blood pressure, daily steps, and calories.
Environment
BME680 & STM32 sensors monitoring temperature, humidity, pressure, air quality (PM 1.0/2.5/10), and gas resistance (MQ).
AI Perception
Pose detection, human following, context awareness, and advanced spatial mapping using RealSense and RPLidar.
About the Project
Final Year Project — UET Lahore
GRACE is a Final Year Project (FYP 2022) from the Department of Mechatronics and Control Engineering at the University of Engineering and Technology, Lahore.
Most student projects try to build a humanoid that barely works. GRACE focused on practical elderly independence — companionship-first robotics with a real-world application.
- Group 3 Members:
- • Muhammad Anss (2022-MC-01)
- • Anas Gulzar (2022-MC-07)
- • Alishba Ramzan (2022-MC-35)
- Department: Mechatronics & Control Engineering
- University: UET Lahore
- Year: FYP 2022
Project Poster
The official FYP poster detailing the core architecture, capabilities, and goals of the GRACE project.
