Robotics Challenge Team Members Tyler Quintana Tyler Gus Josh Cogdill Raul Davila John Augustine Kelty Tobin 1
Robotics Challenge: Team Multidisciplinary: Computer, Electrical, Mechanical Currently split into 3 subsystems/pairs: Controls and Sensors: Tyler Quintana and Tyler Gus Beacon: Josh Cogdill and Raul Davila Chassis/Mechanical: John Augustine and Kelty Tobin 2
Robotics Challenge Tenth annual engineering challenge sponsored by NASA through Colorado Space Grant Consortium held in April Design an autonomous robot to navigate multiple courses at the Great Sand Dunes Natl Park and find a radio beacon, simulating unmanned space exploration Restrictions: 1) Must be autonomous - no remote control or GPS 2) Must remain on the ground - no jumping, launching or flying 3) Must fit into one of two size categories: Division 1: Less than 1.5kg (~3.3lbs) (what we have chosen) Division 2: Less than 4.0kg (~8.8 lbs) 1) Wheelbase may not be longer than 20in and max height may not exceed 30in 2) Maximum $500 budget for final design 3
Sensors: *Note: prototyping does not count towards $500 budget 4
Controls and Sensors: Tyler Quintana & Tyler Gus 5
Planning, Prototyping, & Testing Planning/research: Independent (Kinect, RaspberryPi) Previous Year: Very mechanically focused Arduino, Sharp IR, Ultrasonic Arduino Uno vs. Arduino Mega 2560 Testing Sharp IR sensors: require heavy conditioning of (analog voltage) output Testing MPU-9150: output is very accurate but still requires conditioning 6
Controls and Sensors The brain: Arduino Mega 2560 (master) Beacon signal receiver: Arduino Fio with XBee module (slave) Primary sensors: Sharp infrared sensors x5 Additional sensors: MPU-9150 9-Axis (gyro, accelerometer, and compass) Pushbuttons on a front bumper 7
Next Semester Continue working on sensor input Continue working on integrating with chassis Testing until challenge in April 2016 Refining algorithm 8
Beacon: Josh Cogdill & Raul Davila 9
Xbee Specifications Xbee RF (Radio Frequency) module allow machines to communicate with each other wirelessly Range 300 ft (100 m) Power Consumption Frequency 50 ma @ 3.3 v 2.4 GHz Protocol 802.15.4 Tx Power Max Data Rate Antenna 1 mw (+0dBm) 250 kbps U.FL Connector 10
Transmitter Transmits the direction it is facing (its heading) to the receiver Transmitting Fio takes the heading from a compass and sends it to the transmitting Xbee which in turn sends it to the receiving Xbee Using the HMC5883L Magnometer as a digital compass, which operates in continuous mode 11
Receiver Receives the data (a compass heading) and sends it through serial to whatever device needs that data (in our case the Arduino board) Data is divided by two, making the max value 180, in order to decrease the amount of bits required from transmission The antenna needs to rotate vertically so that the receiver can determine the heading of the beacon relative to the greatest signal strength Without varying the signal like so, the receiver would not be able to determine which direction the beacon was transmitted 12
Transmitting System Receiving System 13
Start Read Heading/RSSI Beacon Detection Save Heading at Strongest RSSI Logic Set Desired Compass Reading to (360-Heading) 180<Reading< 360 0<Reading<18 0 Turn Left until Compass is (Reading+-10) Turn Right until Compass is (Reading +- 10) Proceed Forward End
Beacon Location Algorithm Read heading from beacon at strongest signal strength Robot must travel in opposite direction of beacon heading Calculate 360-heading for a desired robot heading Determine which direction to turn (left/right) depending on the desired robot heading Heading between 0 and 180 turn right, else turn left Continue reading until robot heading is within 10 degrees of the desired heading Path found 15
Looking Forward Construct physical rotating beacon Test implemented code with rotating beacon Various weather conditions Various environments Integrate this code with sensor code on the robot itself Refine code
Chassis/Mechanical: John Augustine & Kelty Tobin 17
Chassis Design Two Motors Custom suspension Sensor, board, display mounts with wire routing Internal belt drive Multi part 3D printed assembly with snap fits and screw mounts (39 components) 18
Board and Sensor Layout Compact design to save weight 19
Wheel Design Primary design constraints Be minimal and light Maintain some rigidity Durability for the expected use Allow for climbing of objects Optimal angle of contact: 90 Materials RC rubber wheels: heavy and expensive 3d printed: loses rigidity and durability while increasing weight at large sizes Foam core: extremely light, may break easily 20
Motor Controls Avoid turning them at desired driving speed initially Allows for establishment of traction Much finer control of robot motion Use the data from sensors to dictate max driving speeds for optimal bite affected by: Angle of surrounding terrain Angle of contact of wheels varies through path motion Coefficient of friction of the wheels 21
Future Chassis and Linkage Design Issues to address Stay underweight: 1.5kg Limit tread motion: currently 5 d.o.f Use Nested ball joints. Must limit to provide optimal contact Avoid binding tread assembly allow for increased accuracy in sensing & movements 22
Robotics Challenge: Questions 23