Traffic Light Romote-control System based on Wireless Sensing Network

Time: Sep.2011-Nov.2012

In this project, we considered two problems at the cross:

1)       The car drivers’ view of the traffic light may be blocked by big trunks ahead of them.

2)       During rush hours, when encounter a red light, there’s no room for normal cars to yield a life way to an ambulance or other special cars.

So we build a remote sensing and controlling system based on Zigbee wireless technology, which allows vehicles to monitor and change the status of the traffic light.

The prototype is showed as follow:

At above is a model of a four-direct traffic light with count-down.

The blue box at bottom is the on-vehicle system.

On-vehicle part with touch screen

It shows the real-time status of the traffic light ahead of you, using Zigbee technology.

Cars like Ambulance or Police cars are given the right to alter the status of the traffic light when emergency happens.

Virtual-Reality(VR) part, developed on Unity3D.

We make the virtual scene interact and work synchronized with the model aforementioned, using zigbee as well. With this we can test the performance of our system without going to the real road. This system can also be used to train drivers.

The whole system.

The picture is took at Jiangsu Province College Students Physics Competition.

Award of the Project

First Prize in Jiangsu Province College Students Physics Competition, Nov. 2012

The National Educational Robots Contest

In November 2011, two of my team members and I participated the The National Educational Robots Contest in Shenzhen University.

It's a pass tracking un-manned vehicle like robot based on microcontrollers. 4 IR sensors are equipped underneath to track the black line on the map.  We intend to control the travel distance and turning angle with encoders on the wheel.
We had two kinds of robot: 51-MCU based and ARM based, and 3 of each. Above is the 51-MCU based one.

It took us over one and half months to make and test these robots.

The video above shows how the robot works. It pushes the colored cellinders toward their coresponding targets.

It was my first time literally working in the lab!

At the opening ceremony (Shenzhen, China).

-

Och! the day before the final race, our robots refused to work in the new environment in Shenzhen.

Our team altered the hotel room into a lab that night. The little robots eventually worked at midnight.

-

above are took right at the competition day.

All sensors were required to be disarmed before the race, we had to equip them on scene.

Any small mistake could lead to disaster, since we had to win as a team.

-

At the beginning  of the project, I was so lack of some basic practical knowledge, that I had to almost learn from the scratch. Luckily, I had good teachers, I couldn’t appreciate more of their help.

This is the first practical projected I had ever participated, I think it taught me more than the 2 years regular courses did. Once we took on this job, we had to make our most to reach the best outcome.

When we finally stood out among competitors, I felt my effort paid back!

The Indoor Positioning System——My Graduation Project

Time:Dec.2012-June.2013

The aim of this project is to develop a 3D indoor real-time positioning system used on fast moving objects like MAVs(micro air vehicles). The positioning information can be demonstrated on an upper computer in real time in an easy-to-read way. All the communication and data transferring are accomplished wirelessly.

It’s part of our indoor quadrotor program. The whole program was inspired by the MAV researches carried by MIT and UPenn.

Technologies and tools involved in this project

Msp430, Altium Designer, Labview, Zigbee, Ultrasonic generator and detector, CCS, GPS, 433 communication module.

My positioning method is to deploy a group of reference nodes inside a certain space, and measure its distance to the moving node. Each of them contains a microcontroller, a Zigbee wireless communication module and sensors for distance measuring. With the extension of space, the number of node increases, so each node sshould carry the traits of low cost, low power consumption, and fast speed.

Each reference node grasps the signal from the moving node, measures the distance between them, and transfers the distance information to a central PC. I put the 3D position calculation work on the central PC to utilize its advantage in speed over microcontrollers.

The reference and the moving node

Reference Node

MSP430 G2433 microcontroller. MSP430 serial, with its reputation on ultralow power consumption and cost of less than 2 bucks.

Zigbee module. Long distance, capable of self-organizing, with maximum of 65530 nodes in a wireless network.

Ultrasonic detector.

433 receiver. An off-the-shelf module, to compose the distance measuring part. In pair with the transmitter.

Moving Node

MSP430 G2433.

Zigbee module.

Ultrasonic transmitter(top). Emit a 60 degree ultrasound sphere so that detectors on different directions can receive the signal.

433 transmitter. 

The moving node is designed to be light-weight. I designed a PCB with Altium Designer to enclose all the components, make the little node looks neat and compact, easy to be mounted on MAVs as well as other moving robots and objects.

The PC controller

PC Interface

Zigbee with a serial to USB board, the PC just need a USB port to communicate with the network.

User Interface

I wrote this User interface with Labview, and G-Language provided by NI. It can stop and run the positioning process. The 3D coordinates will be displayed in number. I kept the raw data from nodes on display for inspect.