Build and control a Python-powered robot.
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
Title |
| |
Title of Series | ||
Part Number | 08 | |
Number of Parts | 169 | |
Author | ||
License | CC Attribution - NonCommercial - ShareAlike 3.0 Unported: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this | |
Identifiers | 10.5446/21109 (DOI) | |
Publisher | ||
Release Date | ||
Language |
Content Metadata
Subject Area | ||
Genre | ||
Abstract |
|
00:00
Student's t-testPower (physics)RoboticsGroup actionMultiplication signMobile WebGame controllerClosed setLecture/Conference
00:38
Line (geometry)Pattern languageProgrammer (hardware)Projective planeTheoryPlanningLecture/Conference
01:10
Streaming mediaComputer hardwareDevice driverUltrasoundPower (physics)Game controllerServer (computing)Network socketGraphical user interfaceTelecommunicationCommunications protocolGotcha <Informatik>DistanceUltrasoundPower (physics)Computer hardwareHydraulic motorCommunications protocolRoboticsGame controllerRange (statistics)Multiplication signStreaming mediaDistanceMedical imagingSoftwareFrame problemGreatest elementServer (computing)Network socketPort scannerScaling (geometry)Endliche ModelltheorieChemical equationSet (mathematics)FrequencyDemosceneTheoryVotingRule of inferencePhysical systemRankingFlow separationVideo gameBasis <Mathematik>Field (computer science)Wave packetSurfaceAreaComputer animation
05:20
Communications protocolTelecommunicationOpen setData managementComputerOpen sourceObject (grammar)Machine visionLibrary (computing)Medical imagingLecture/ConferenceComputer animation
06:20
Type theoryPredictabilityEndliche ModelltheorieVirtual machineOpen setLecture/Conference
06:44
PredictionData modelPairwise comparisonFrame problemMachine learningWave packetComputer-generated imageryReliefComputer fileDefault (computer science)Maxima and minimaKey (cryptography)Control flowEndliche ModelltheorieFrame problemScaling (geometry)Medical imagingRectangleFerry CorstenAlgorithmSimilarity (geometry)Object (grammar)Position operatorOpen setClique-widthMultiplication signLink (knot theory)Software maintenanceJSONXML
09:25
2 (number)Medical imagingThermodynamic equilibriumLecture/Conference
09:56
Computer-generated imageryHypermediaLine (geometry)Interface (computing)Execution unitPoint (geometry)Video gameTheory of relativityUsabilityAsynchronous Transfer ModeCodeMaxima and minimaResultant2 (number)Computer animation
12:02
Process (computing)Set (mathematics)RoboticsSynchronizationNetwork socketServer (computing)Communications protocolLecture/Conference
12:23
Server (computing)Group actionMoment (mathematics)Scale (map)Frame problemFirst-person shooterImage resolutionComputer-generated imageryServer (computing)SynchronizationConnected spaceMultiplicationClient (computing)Network socketMoment (mathematics)MomentumVideoconferencingMedical imagingEndliche ModelltheorieImage resolutionFlow separationFrame problemObservational studyDataflowInequality (mathematics)Graph coloringVector spaceGame controllerMessage passingGroup actionComputer hardwareSerial portLecture/ConferenceComputer animation
15:38
Scale (map)Image resolutionFrame problemFirst-person shooterComputer-generated imageryNormal (geometry)TheoryCoefficient of variationFrequencyEndliche ModelltheorieDataflowSoftware developerDemo (music)JoystickRoboticsPower (physics)UltrasoundLecture/ConferenceComputer animation
18:47
FamilyWordData recoveryEndliche ModelltheorieSequenceContent (media)Multiplication signFunction (mathematics)RoboticsHydraulic motorJoystickStreaming media
21:58
Scale (map)Image resolutionFrame problemFirst-person shooterComputer-generated imageryComputer animation
23:35
Moment (mathematics)Utility software
24:03
Lecture/Conference
24:24
Moment (mathematics)
25:02
Moment (mathematics)Lecture/Conference
Transcript: English(auto-generated)
00:00
Introduce you Antonio Spadaro and he's going to talk about how to micro and how to Python power a robot So, please big hand for Antonio
00:29
Begin with something about me He goes cool students the next year I'm an interest here. I introduced to fight line Linux user and a pattern programmer
00:48
I Live yeah, I live in Italy and Italy and from Okay
01:05
Small project because it's too easy to to make This robot what do you do? This robot to stream the camera you take a camera
01:21
By camera exactly It don't have to hit against to the wall thanks to ultrasonic sensor It Recognize and watch humans so one with the camera
01:43
Where it is an human it model the camera and see The human so what is the hardware? The hardware is a raspberry p3
02:02
a Pappy camera abuse For motor DC to motor driver for control the motor DC To servo motors An ultrasound sensor and a power bank for power
02:26
This is
02:49
This is a little Scan of food a hardware at the center. We thought we found there are very petri
03:04
Bottom away from where we can found a picamera and a servo motor The picamera send the data for a very petri Oh Dr. Mob the servo motor
03:20
The ultrasound sensor There's a repeat radio from ultrasound sensor the data and control the motor for do not to hear low against the wall a PC control All motors of a therapy and a servo motors
03:45
The software is a socket server on the therapy camera socket server To go For control the robot and for receive the
04:06
frame of the camera and The server command the motor the motors server motors and the camera
04:23
Undirectly it to control the server motors from the camera and The motors from the ultrasound sensor The communication protocol protocol user they use Wi-Fi
04:42
But why Wi-Fi wind or not I use Bluetooth Wi-Fi is more speed is a more range of distance and We can connect to more days
05:03
So we can send Quality image No time I can We can
05:22
Go Open CV these are awesome library for recognize a human recognize a human
05:48
object It is an open Acronym of open source computer vision this library allowed to
06:02
Use To manage a day image from PC But oh it work. Oh it Recognize object the humans the animals first they made you
06:25
Will be Will be signed from a machine learning on Open CV That trying the model and return a predictive model
06:42
hard-cached This predictive model will be compared frame We will compare frame with this predictive model and it return the Position for the object before it found it. This is a little
07:05
Little example of how to work First we import Open CV Then we open the camera for the PC we connect of the camera, then we load
07:23
the arcade the predictive model We File We capture the frame We make a gray scale of the image
07:44
Well search some face on the frame It this face return on numpy array with To the position of the face and the
08:04
width and the height and Now we make a rectangle around it around it Then we will show the image with the rectangle and all
08:22
And we found before we have pressed exit and it exit This is the result. It recognizes smile, people, patternions and the drawing
08:42
Because it use a simple algorithm that to try to recognize a similarity from this This is the first from us on our very picturey for found a
09:05
face on an image of 60,000 40 for 4080 We it
09:20
Will be Zero 52 seconds zero 52 seconds on 30,000 20 for 20,000
09:40
40 image it will be 0 5 10 15 seconds But now we can optimize it
10:00
We can insert a simple line of code On this line we Specifically Minimizes of the face and the maxes of the face This is the result of the performance
10:26
Instead of zero 0.52 0.52 seconds it will be 0.17
10:40
0.17 0.17 instead of 0.15 0.15 it will be 0.04 0.04 0.04 0.08 1.08
11:01
1.08 1.08 2.08 3.08 4.08
11:22
5.08 6.08 8.08 But it is more small, RSVP is more small, because if we take an Arduino, we insert Wi-Fi module, camera module, shield or some shield, it will be more big of RSVP.
11:57
So I use RSVP instead of Arduino.
12:15
A synchronous socket server. For command the robot, I use a socket server with a SYNGIO protocol.
12:24
This socket server allows a synchronous connection for multiple clients. Synchronous action on the same moment, we can move a servo motor while we control the motor.
12:44
And we can control the hardware with this.
13:09
For syndrome, the frame from RSVP to client, I use another socket server.
13:20
But the performance is a problem, a little problem. The big resolution has a big frame delay. If we want to send 60,000 for 4080 image, we can found the video at 4 FPS.
13:51
But the grayscale is more speed. So if we want a fluid video, we should use a lower resolution with grayscale.
14:12
Instead, if we want the quality with the colors, we should use a higher resolution, but it will be slow.
14:30
So because the color has three channels of color, instead of black and gray, it has only one color channel.
14:51
So the RSVP use, when we start the robot, RSVP make an hotspot, a Wi-Fi hotspot.
15:03
And the client connect to the Wi-Fi hotspot, and then we can control it. For connecting the three cameras, I use a visualized CSI camera socket interconnection.
15:35
Camera serial interconnection, sorry. And later, it is more speed of an USB, a normal USB port.
15:50
Okay, with the two servo motors, we will move the camera for the face.
16:16
So I could be a speed demo or not.
16:27
Yes, this is the robot, the little robot.
17:22
So at the moment, the RSVP, this is the power bank that power the RSVP and the motors, and this is the ultrasound sensor.
18:07
Okay, now it power on.
18:42
So for controlling it, I use a joystick gamepad. Now, if you have a PC, you can see a new Wi-Fi hotspot, a led robot.
19:36
Okay, now I'm connected to the robot.
20:11
We can move the robot, the servo motors with the joystick.
20:23
We can rotate the robot, and if we want, wait a moment, we can control the robot with a simple joystick. We can, it can go back, can go back, forward, can rotate on itself.
21:05
We can enable the camera, now I can see the output of the camera.
21:36
Okay, this is the streaming.
23:41
At the moment, it is at the minimum quality.
24:03
We have white and black, it is more speed, but white and black. We can set the more beautiful quality, but it will be not lower.
24:21
And with the color, so if somebody want to try it, Robin, thank you.
25:14
If somebody, anybody has questions, like a moment.