"Follow Me" Robot for Assisted Living

2017 ELE Engineering Design Project (NMK02)

---

Faculty Lab Coordinator

Naimul Mefraz Khan

Topic Category

Intelligent Instrumentation

Preamble

Assisted living is an important aspect of life for the elderly. One of the critical problems with assisted living is instant access to important things, such as water and medication. A smart robot that can follow elderly person everywhere in the house automatically through visual processing can be very useful in that regard. Especially visual processing can provide additional functionalities, such as detection of medical emergencies by detecting whether the person has fallen down.

Objective

To build a smart robot (car) that can automatically follow a person around in a room through image processing. The robot will receive data from a Kinect, and through the Kinect API, assess if the person has moved away from the robot. Once the robot detects where the person is, it can move towards the person, and stop when the distance has reached a minimum threshold.

Partial Specifications

The robot car should:
1. Assess the position of the person with respect to the car through the depth image obtained by Kinect.
3. Calculate the required distance and turning angle to reach close to the person.
4. Move towards the person and stop when the distance has reached a minimum threshold. The closeness to the person will be determined through continuous processing of the the depth image obtained by the Kinect.

Suggested Approach

1. A 4-wheel car can be built, or an existing RC-car can be modified so that it can be controlled through an Arduino microcontroller or similar.
2. A small laptop with a Kinect/similar depth sensor connected to it can be mounted on top of the car.
3. A voltage regulator circuit should be built so that the Kinect can be battery-powered.
4. The depth-image processing software will reside on the laptop.
5. The distance assessment and required turning angle and speed commands will be communicated to the car through the microcontroller.

Group Responsibilities

1. Study the required hardware (RC cars, Arduino, Kinect). 2. Study the required software APIs (Kinect). 3. Design the car rig that can fit all the required components. 4. Program the depth image sensing and distance assessment. 5. Test the demonstrable product under different conditions, both idea and non-ideal, and report limitations.

Student A Responsibilities

TBD

Student B Responsibilities

TBD

Student C Responsibilities

TBD

Course Co-requisites