Robotics

how to interpret sensor data into a mathematical form readable by computers

There are only 3 steps you need to follow: Gather Sensor Data (data logging) Graph Sensor Data Generate Line Equation

Some sensors (such as sonar and Sharp IR) do not work properly at very close range

pseudocode: read left_photoresistor read right_photoresistor if left_photoresistor detects more light than right_photoresistor then turn robot left if right_photoresistor detects more light than left_photoresistor then turn robot right if right_photoresistor detects about the same as left_photoresistor then robot goes straight loop

Photovore Algorithm, Improved This algorithm does the same as the original, but instead of case-based it works under a more advanced Fuzzy Logic control algorithm. Your robot will no longer just have the three modes of turn left, turn right, and go forward. Instead will have commands like 'turn left by 10 degrees' or 'turn right really fast', and with no additional lines of code! pseudocode: read left_photoresistor read right_photoresistor left_motor = (left_photoresistor - right_photoresistor) * arbitrary_constant right_motor = (right_photoresistor - left_photoresistor) * arbitrary_constant loop

stimulus-response

a plan of action

combine the best of both Reactive and Deliberative control

Droidmakr (Cliff Boerema) came up with an interesting idea for a light-tracking head with a form of peripheral vision. As often happens, the circuit turned into something different -- a photopopper:

All done with a single 74HC14 (the '240 being a motor driver).

I tried the same setup with the 74*240 (with an extra inverter per motor) and 7404, but the 74HC14 seems to work best.

I wanted to share an adaptation of the Schead v4, that I have been experimenting with. It is (for lack of a better term) a Master/Slave Schmitt Comparitor Head (M/S SC-H). With the addition of a 74 AC 240 or two (as motor drivers) and a pair of motors, you can put together an interesting little light seeking, wheeled robot with peripheral vision.

As long as the light reaching the photo-bridge of the Master SC-H is balanced, then the Slave SC-H acts as a regular, lone SC-H would. So, if one of the slave photo-diodes detects more light then the other, the inverter that controls the motor on that side changes states and is now the same as the inverter of the Master SC-H tied to the same motor. This turns that motor off and the robot will pivot around the stopped wheel toward the greater light source until the light on each sensors is balanced and the motor again begins to turn.

I am also using SCar to continue experimenting with Stacking separate Sensor/Behavior circuits onto a robot. I will post more as progress is made.

It is powered with two 3.3F Goldcaps. They can be charged in a few seconds. When they are charged, MAXIBUg gets "afraid" of light, and wanders of to go to play "in the dark". After a while, about 20 seconds (depending on the current used by the two motors ), the power has dropped, and it wants to "eat". It gets light attracted, and will turn and go to the light. When it gets there, it will recharge and still will be atrackted to the light until it reaches a trigger voltage , at which it gets "afraid"of the light again. This will go on all day until someone turns off the lightsource. While doing all this it also will backup when bumping into something.

Because of the "on-off" output of the first schmitt trigger, the inputs for the LDRs will switch. That's why it gets light atracted -light afraid. This also means that you cannot use IR diodes (like SHF205). You have to use LDRs !

This adapted photodiode is not as sensitive as large area types so C2 may need to be reduced to 0.01uF while the value of R2 and R3 can be increased by a factor of 10.

Two leaded phototransistors can also be used but may require extra shielding to reduce light current in the bridge to acceptable levels

basic photopopper functions plus reverse -- all on a single chip

I will go through the process of making a guitar string physical touch sensor.

看到上面一个大大的玻璃试管和里面的液体、引脚我就知道肯定是通过液体摆动跟触角接触面积引起的阻值变化测量角度的

The current through a photodiode is directly proportional to the light intensity

The photodiode and phototransistor can be both photovoltaic (generators of potential difference) and photoconductive (modifiers of an electric current), depending on the application.

A reverse-biased photodiode operates in what is called photoconductive mode, since the conduction of the semiconductor junction varies with the illuminating light intensity.  If the reverse-biased voltage is relatively large (i.e. several volts) the reverse-biased photodiode will have a very fast response time (much faster than an LDR) and is suitable for detecting light signals that vary down to a time scale of a fraction of a microsecond.

The Sharp IR Range Finder works by the process of triangulation. A pulse of light (wavelength range of 850nm +/-70nm) is emitted and then reflected back (or not reflected at all). When the light returns it comes back at an angle that is dependent on the distance of the reflecting object. Triangulation works by detecting this reflected beam angle - by knowing the angle, distance can then be determined.

The IR range finder reciever has a special precision lens that transmits the reflected light onto an enclosed linear CCD array based on the triangulation angle.

The Sharp IR has a non-linear output. This means that as the distance increases linearly (by set increments), the analog output increases/decreases non-linearly.

Usually, the operating system takes all the values from the sensors, averages them, then finds the one that is the furthest away from that average

The Ant Farm is built next to a very large window that faces east. As a result, with the lab lights off, the brightest light source is always to the east. The robots can use this as a reference and then find all the other directions.

This chip is great for building robots because it has extensive input/output hardware built right into the chip. This reduces the need for external components.

These bots are powered by a Gold Cap and for a period of about one minute they move, always looking for the brightest lightspot, so in fact they will even follow a lightsource.

All these bots are powered by a 3,3F Gold Cap ( F= farad). You can charge them with a regulated power supply

the two 5 mm red LED's it is capable of following a light source.

One essential difference is that the Nv responds immediately to an input, and sends the output for a time duration -- the delay occurs AFTER the output is sent. The Nu responds to an input after a delay and sends the output continuously -- the delay occurs BEFORE the output is sent.

"on" first, then a delay, then "off"

delay, then "on", stays "on"

By just tying a neuron's bias resistor to Vcc, rather than to ground, you can make a "regular logic" (active high) Nv:

putting diodes and other resistors in parallel to give different charge vs. discharge rates

>