Sunday, July 24, 2011
A remote-controlled robot relies 100% upon the operators senses. For an autonomous or semi-autonomous robot, we need to impart awareness of its environment. Even for a robot that is not autonomous, telemetry of sensor data can be valuable. In this blog, I surveyed the low-hanging fruit of the abundance of inexpensive sensors. The average cost per sensor is $12. Almost anything you can imagine can be measured: proximity, distance, temperature, light, magnetism, sound, barometric pressure, vibration, etc. I am really tempted to combine a bunch of sensors and construct a tricorder (like on Star Trek) for the Ghost Adventures show. (Can a spirit actuate a capacitive touch sensor?)
Sensor survey matrix: [Link]
My other stuff (kayaking, plant ID, sea beans, etc.): [Link]
Saturday, July 23, 2011
Building robots is a fun activity; but for me there are also high ideals that guide my efforts. Helping others to succeed, encouraging experimentation, sharing knowledge, making learning fun, building friendships and inspiring a community of enthusiasts are all desires that focus my passion. Four inter-related disciplines that are foundational for this effort are summed up in the acronym RCOC:
Repeatability: this blog exists to serve other robot enthusiasts. That means I will try to refreign from using odd, one-of-a-kind or hard-to-obtain parts. Any person of average intelligence and strong desire should be able to reproduce anything posted here. I will do my best to assist anyone having technical issues.
C.O.T.S. (Commercial Off-The-Shelf (products)): robots will be built using generic parts from major vendors where possible (like McMaster Carr, Digi-Key, SparkFun, Pololu, Servo-City). Ideally, each part or a similar one should be available from multiple sources.
Open Source: everything is disclosed: electrical, mechanical drawings, source code and parts lists to enrich the community. That also means that everyone is encouraged to contribute to the knowledge base.
Competition: to promote creativity and excellence, non-destructive competitions will take place that will cultivate more robust, flexible and effective designs.
If you want to be a part of this community, contact me and join the E-mailing list (and not miss postings).
James the Quack
My other stuff (kayaking, plant ID, sea beans, etc.) [Link]
A brushed motor without gear reduction is practically useless for robotics. To move things (perform mechanical work) we could use gear-head brush motors, servos or stepping motors (yes, there's others...solenoids, piezo thing-a-ma-jigs, etc.) For locomotion, high-torque and 'horse-power', nothing beats a good gear-head motor. The gear reduction effectively multiplies the torque of the original motor, trading speed for 'umph'. The link below takes you to a matrix of the motors that I have evaluated so far.
Motor Matrix: [Link]
My other stuff (plant ID, kayaking, sea beans, etc.) [Link]
When I was playing with 'Jim-Bob', my toy tank-tread robot, I was appalled by how quickly those little gear motors ate AA batteries! I hate throwing AA alkaline batteries in the trash for so many reasons: environmental, financial and personal thrift. The weight of my first robot, 'Little Jimmy' was nearly 50% gel cell battery. The electric car industry faces similar dilemmas of achieving the needed voltage and capacity at the lowest possible weight and volume (bulk) at a cost people can accept. There are a lot of choices in battery chemistry for secondary (rechargeable) cells: lead acid, gel cells, nickel cadmium, nickel metal hydride, lithium-ion polymer. They do not come without risk: a few years ago the industry had a rash of exploding lithium lap top batteries. Each type of battery chemistry requires a different charging system. (Batteries in the hands of idiots are dangerous!) Every element of robot design and construction requires research, and batteries are no exception. Attached is a matrix of my research on various battery specs. The conclusions I came up with is that for an Ant-Weight (1 lb) robot like 'Jim-Bob', a 6V NiMH battery pack is a good choice. For a Hobby-Class robot like 'Little Jimmy' (10 lbs) a lead acid gel cell is just fine.
Battery Matrix [Link]
My other interests (Kayaking, plant ID, sea beans, etc. ) [Link]
Monday, July 11, 2011
From left to right top row: RC transmitter, RC receiver, XBEE, XBEE Breakout Board
From left to right bottom row: XBEE USB Explorer, Wixel, Nordic, BlueTooth
Comparison (check out all 3 page tabs listed at bottom of spreadsheet) [Link]
Though I have been using an IR (infra-red) remote control for indoor prototype robots, RF is required for good distance and performance outdoors. Due to outrageous Chinese exports, the cheapest and easiest system to implement is the conventional RC (radio control) using PWM signals. Any of the other 2.4GHz link modules will require more smarts in the robot to process the commands, and either a computer or an elaborate self-made transmitter control box. The one big advantage of using these RF module links is that they are transceivers (i.e. they both transmit and receive data), so they make telemetry possible (the receiving of sensor and status signals form the robot.)
My main blog list : [Link]
Saturday, July 9, 2011
Tuesday, July 5, 2011
Saturday, July 2, 2011
In this blog I explore using a universal TV remote control to send IR signals to a PICAXE-18M2 micro-controller to command a servo, LED and a gear-head motor to actuate. This demonstrates how a robot could be remote controlled easily and inexpensively.
My Instructable on the circuit:
My video of the demo:
My other stuff: