Reproduce demodulated RF signal to control the car Remote Control Sensor Car for Vehicle to Vehicle Communication Testing Julian Santorelli, Robert Morawski and Tho Le-Ngoc Department of Electrical & Computer Engineering, Broadband Communications Research Lab, McGill University Motor vehicle accidents take thousands of lives a year in Canada alone Several studies have shown up to 60% of these crashes could be avoided if the driver was warned just a half second sooner[1] Large number of such crashes are due to driver factors[1]. Drivers suffer from perception limitations, resulting in large delays propagating emergency warnings Short term Goal: Create a prototype for a sensor car. Equipped with ultrasonic sensor and Bluetooth that can be controlled remotely by the user Long Term Goal: Implement a mesh network with multiple sensor cars to test Vehicle to Vehicle(V2V) communication in a controlled environment Wireless Communication ZigBee communication protocol was used. It is Based on the IEEE 802 standard. Used for low power, short range transmission with relatively low data rates. Designed for long battery life. •ZigBee Coordinator - Selects channel parameters, allows nodes to join, can route data, has to be awake •ZigBee Router- Must join network before routing data, can allow other nodes to join, always awake •ZigBee End device – Transmit or receive data from parent only, can sleep Mesh network: Each node must be able to transmit and receive data from every other node ♦ Results Waspmote board and gateway by Libelium Xbee ZigBee RF modules by DIGI International inc RC Car and Ultrasonic sensor ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ ♦ Materials & Methods Figure 2: Image taken from[3] Figure 3: Series 2 and Series 2 Pro , taken from [4] Study PCB To interface Sensor node with RC car Reproduce output For full control of car via microcontroller Create Network Set up Waspmote device as a mesh network Calibrate sensors Send raw data back from car sensors to the gateway Create user control On screen GUI or via Accelerometer remote control Interface between Sensor Node and RC Car Figure 5 :PWM Signal, taken from [6] Simple one wire interface. Feed the reproduced signal from Waspmote board’s digital output directly to the IC on the car’s PCB 1,2,3 are constants 5 controls speed 4 controls turning 6 is fixed length of signal Easy implementation and easy to repeat over many cars No need to add or remove hardware on PCB Ford is currently testing V2V communication using NFC and WiFi as part of its intelligent cars program. This System would give drivers advanced warning on upcoming accidents, warn drivers if there is an oncoming car at a difficult or obstructed intersection or let them know of sudden changes in speed from cars in front of them. Set up Waspmote device in a Mesh topology with the corresponding node types Figure 7: Network set-up Network Creation ZigBee Router ZigBee Coordinator ZigBee Router Program network joining parameters on Xbee modules; Channel, PANID, stack profile etc. Coordinator scans nearby nodes and sends out message for the corresponding nodes to join Once the network is set-up information can flow from computer(the gateway) to either of the routers. As well, each router can route the message to the indented device if it is out of range Smart routing handled by Waspmote API Figure 8 : Network parameters displayed with X-CTU Software ZigBee Routers are used so cars can route warnings or messages they receive to the next node Figure 9 : Routing protocol employed by Waspmote API, Taken from [7] User Control Accelerometer Samples Accelerometer Values at 160 Hz and performs minor filtering via the API Converts the larger Accelerometer values, both X and Y, to a smaller subset of values Sends data at very frequent rate to the other Waspmote, Waspmote processes small subset of values via a lookup table kept on the flash memory, then carries out associated instruction On Screen Loads a user chosen map(to scale) and creates ½’ by ½’ grid Sets up serial communication with Waspmote Gateway Once a path is drawn, it is analyzed, broken down into a subset of very simple commands, formatted to fit Waspmote’s packet design, then sent serially to the gateway Once the packet is sent, the Waspmote on the car processes the commands and executes them in succession ♦ Future Application V2V communication trials are going on in 6 major cities in the US, headed by their Department of Transportation. Such systems when implemented can reduce up to 80% of unimpaired accident 7 and can even relieve traffic congestion as well as pollution by offering real-time traffic and alternate routes Limitations and issues Need to constantly drive and produce a PWM signal to control car, requiring heavy use of the network if real time Future Work On screen GUI control to work in real time, both sending data and receiving data from the sensor node. This would require heavier processing needing an external application Create multiple sensor cars and test communication of the car’s data between them, test if cars can avoid each other based on position and ultrasonic data ♦ Conclusion [1] C. D. Wang and J. P. Thompson. Apparatus and method fo motion detection and tracking of objects in a region for collision avoidance utilizing a real-time adaptive probabilistic neural network, 1997. US.Patent No. 5,613,039. [2] Robert Poor, Ember Corp., Wireless Mesh Networks, http://www.sensorsmag.com/networking-communications/standards-protocols/wireless-mesh-networks- 968,2003 [3] http://www.libelium.com/products/waspmote [4] http://www.digi.com/products/wireless-wired-embedded-solutions/zigbee-rf-modules/zigbee-mesh-module/xbee-zb-module#overview [5] http://www.geeky-gadgets.com/geeky-toys-the-rc-spy-car/ [6] Chris Leung, Remote Control Car Interface Solutions, 2008 [7] http://portal.ember.com/node/682 [8] http://electronicsbus.com/vehicle-to-vehicle-wireless-communication-system-wifi-nfc/ [9] Viknesh Vijayenthiran. Six Cities Named For New Vehicle-To-Vehicle (V2V) Communications Trials. http://www.motorauthority.com/news/1060101_six-cities- named-for-new-vehicle-to-vehicle-v2v-communications-trials. 2011, ♦ References Figure 1: Mesh network [2] Figure 4: Ultrasonic sensor and RC car Taken from [3] Taken from [3] Taken from [5] Figure 6: One wire interface to PCB Figure 10: Accelerometer control Figure 11: Gui for Pc control Figure13: Ford V2V assistance, taken from [8] Taken from [9] ♦ INTRODUCTION