Smart Shopping Cart ( Trolley ) Project is conducted the prevail automation and make people’s life easier. This trolley has different abilities including tracking people, gesture recognition, picking up and placing objects with a 3DOF robotic arm and object recognition with RFID tags and IOT technology. The Kinect sensor has been used in this project in order to capture images. This trolley can be used in dangerous places and places where are harmful to human health in order to decrease human being in such places.
Step 1: Designing the Cart
This cart is designed to carry up to 12 KG items in it with the average speed of 1 meter per second. A robotic arm is designed for this cart to pick up objects and put them in the cart's basket up to 200 grams. Choosing the correct motors and location of them is a crucial task as the cart does not have a symmetric shape. We used SolidWorks to design the first idea.
Step 2: Tracking People : the Kinect Sensor
This cart is able to follow people using Kinect sensor through different gestures. We detect different joints of a human body and by measuring the angle of different joints we can recognize different human body positions. As you can see in the videos, If other people come in and out of the camera scope, the Kinect still works well and won't lose the target.
Unlike regular 2D cameras, the Kinect sensor is able to measure the depth of an image, so we can find the distance between a person and the cart. Using this distance we can keep the cart in a certain range of distance from the person. We used PID controller to control the motors speed to manage this distance.
Step 3: 3DOF ARM
This cart has a robotic arm with 3 degrees of freedom in order to pick up objects up and put them in the basket. The arm firstly designed in SolidWorks and then implemented using links, servo motors and a gripper.
The cart stops moving when the detected person raises her hands up. The arm starts if the cart is in the stop state and as you keep your arms straight right perpendicular to your body.
Step 4: IOT: Detecting Items Using RFID Tags
After picking up objects, we read them by RFID scanner which is installed on the cart using RFID tags which are stuck on the items. Then the data is being sent to a data center and a bill can be reported containing which objects were scanned. In the following movie, objects are being scanned and the data is being sent to the data center via wifi.