340 likes | 492 Views
Robotic Concepts. Mehdi Ghayoumi MSB rm 132 mghayoum@kent.edu Ofc hr : Thur , 9:30-11:30a. Robotic Concepts. Announcements: Your final exam will be at Wednesday: Dec-03-14. Nov- 3 (G1,G2 and G3) and 5 (G4 and G5) Lab project presentation.
E N D
Robotic Concepts Mehdi Ghayoumi MSB rm 132 mghayoum@kent.edu Ofchr: Thur, 9:30-11:30a
Robotic Concepts Announcements: Your final exam will be at Wednesday: Dec-03-14. Nov- 3 (G1,G2 and G3) and 5 (G4 and G5) Lab project presentation. Nov- 10 (G1,G2 and G3) and 12 (G4 and G5) Paper presentation. Hw 4 is available due to Monday Sep-15
Robotic Concepts Projects: 1.Radio-Controlled Airplane 2.The robot that predict weather Station 3.The Robot with home security system 4. RasRobot with face recognition 5. Implementing your paper on your own Robot
Robotic Concepts • Projects: • 1.Radio-Controlled Airplane • Raspberry Pi (with camera) • GPS receiver (https://www.adafruit.com/products/746) • Antenna for receiver (optional) (https://www.adafruit.com/products/851) • Medium-sized RC airplane • RC battery and 5V regulator to power the Pi
Robotic Concepts Projects: 1.Radio-Controlled Airplane
Robotic Concepts • Projects: • 2.The robot that predict weather Station: • Raspberry Pi and power adapter • Digital compass/magnetometer (https://www.sparkfun.com/products/10530) • Optical shaft encoder (http://www.vexrobotics.com/276-2156.html) • Barometric pressure sensor (https://www.adafruit.com/products/1603) • Digital thermometer (https://www.adafruit.com/products/1638) • Small breadboard (https://www.sparkfun.com/products/9567) • Square shaft from hardware store • Pinwheel or similar fan-like device • Thin wooden plate • PVC pipe with cap, about 12 inches long, 1-2 inches in diameter • Miscellaneous jumper wires, glue, screws
Robotic Concepts Projects:
Robotic Concepts • Projects: • 3.The Robot with home security system • A Raspberry Pi (obviously) and power adapter • A wireless USB dongle if you want to go wireless with your Pi • Raspberry Pi camera module • Pressure switch (such as http://bit.ly/Owc8aN, for example) • Magnetic sensor (such as http://bit.ly/1cis71c, for example) • Motion sensor (such as http://bit.ly/1c35pQp, for example) • Reed switch (such as http://bit.ly/1k6n2RM, for example) • Large spool of Ethernet cable (crimped ends not required—buy in bulk to save money) • Solder, soldering iron, and miscellaneous jumper wires and connectors
Robotic Concepts Projects: 4. RasRobotwith face recognition
Robotic Concepts Projects: 5. Implementing your paper on your own Robot 1. Object Recognition and Path-Drawing for a Mobile Robot 2. Applying Image Processing Techniques on a Ball Collecting Robot 3. Implementation of Mind Control Robot 4. Implementation of Shortest path planning algorithm without track using FPGA Robot: A New approach 5. The Robot’s Eye Expression for Imitating Human Facial Expression
A credit card sized PC Plugs into a TV or monitor Cheap (ish) ~£25 each Over 1 million units shipped Uses Broadcom BCM2835 Designed for education! Robotic Concepts What is a raspberry Pi?
Robotic Concepts Audio USB Video GPIO Ethernet SD HDMI CPU Power
Robotic Concepts Path Finding Path finding vs. trajectory optimization, local vs. global, Dijkstra, Probabilistic Roadmaps, Rapidly Exploring Random Trees
Robotic Concepts Consider the following situation: Straight Path Trial & Error Path Target Pre-Planned Path Lake or Other Hazard Robot
Several knowledge-based mobile robots and techniques exist Basic idea: If information about the environment is well-known Pre-planning a reasonable path based on known data Adjustments can be made (if needed) along the way If no or little information is available about the environment Possible to collect information and store along the way Problems arise with collecting accurate information Disadvantage: System highly reliant on accurate and detailed data on surroundings Robotic Concepts Knowledge-Based Path-Finding
Robotic Concepts Path Finding Dijkstra,
Works on both directed and undirected graphs. However, all edges must have nonnegative weights. Approach: Greedy Input: Weighted graph G={E,V} and source vertex v∈V, such that all edge weights are nonnegative Output: Lengths of shortest paths (or the shortest paths themselves) from a given source vertexv∈V to all other vertices Robotic Concepts