|Sent on:||Wednesday, October 30, 2013 11:49 AM|
Hello fellow Philly Roboticists,
The November meetup is scheduled for the 21st. Let me know if you have anything you’d like to discuss or demo.
We continued our trend of garnering sizable attendance numbers into the September meeting. Below are my notes about what went down. Please let us know if you have anything to add.
- Earl had a nice talk about “Semiconductors, The Good, The Bad, and the Ugly”. He was able to draw on his years of experience working in the field to give us the inside scoop on the semiconductor industry. He explained how the industry moved from tubes to transistors and diodes, how it produces these devices, conducts quality control, how parts are sourced, and even a little on the economics of these components.
- Mike is starting a new project involving a biped walking robot. He’s basing his design on a number of interesting principles, including: human physiology (e.g. the way our muscles are arranged in push-pull pairs; the way we fling our limbs out then catch them (just watch a baby walk)); passive dynamics (efficiently using the dynamics of the limbs instead of continually driving them. Pioneered by Tad McGeer who had a machine that would walk down an incline without any motors or control systems http://en.wikipedia.org/wiki/Passive_dynamics); and underactuated systems (e.g. having more degrees of freedom than actuators, which is true of just about all walking, flying, and swimming machines or animals. I posted about this earlier http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-832-underactuated-robotics-spring-2009/). He will try to minimize the number of actuators and exploit the natural periodic rhythm of walking. He also desires to minimize the amount of computation and software, which he hopes to accomplish by using some simple timing circuits (like a 555 chip) or (as we learned in the October meeting) FPGA circuits. He might need to use a gyroscope to stabilize the walker, but there may be a way for him to add moving arms to provide “passive” stability. Mike also found some inexpensive but powerful linear actuators that were originally designed to actuate car door locks.
- John mentioned a nice free 3D cad tool for your 3D printing designs and other needs: http://designspark.com/eng/page/mechanical. He also gave us this link to a 3D printer information site http://www.3ders.org//
- Tom & Scott updated the group on their K9 project. They discussed ways of actuating the head and adding sensors. They also want to drive a bunch of blinking LED lights from an Arduino using transistor amplifiers.
- Dave spoke briefly about a recent Tesla coil meeting. He mentioned the MIT Dual Resonant Solid State Tesla Coil kits, which are capable of switching fast enough to actually play music. http://onetesla.com/
- Other links of interest:
Someone had recommended using sensors from SchmartBoard (for Mike’s walker) http://www.schmartboard.com/index.asp?page=products_sensors. Another good source for sensors, processors, and other electronic items that all members of our group should know about is SparkFun https://www.sparkfun.com
Here’s an open source package that can simulate musculoskeletal structures and create dynamic simulations of movement https://simtk.org/home/opensim
I’ll have more to say about the Beaglebone in the October meetup summary (coming to an in-box near you soon!), but I thought I’d share this Derek Molloy video series on the Beaglebone and programming in embedded Linux. http://www.youtube.com/playlist?list=PLF4A1A7E09E5E260A
See you next month!