The servo does not respond as we wanted. Installed Digilent WaveForm to explorer,test and take measurements to see the output from raspberry Pi on the oscilloscop.
Clockwise pulse width of 1.5ms
Counter clockwise pulse width 1,16ms
Konseptet vi kom frem til når det gjelder mekanismen på hvordan best mulig slippe ut godteriet i en kontrolert affere var et slags skovlehjul. Dette gjør at godteriet kommer ut i en bestemt fart og det er også en god måte å unngå at godteriet setter seg fast.
Vi ønsker å andvende en ansiktsgjenkjenning som identifikasjon for brukere i databasen. Databasen vil holde kundens preferanser (forhåndsvalgt) samt tidligere variasjoner av godteri og kundens tilfredshet med disse. Vi må kunne skrive til og fra en lokal database som må kunne legge til nye brukere samt holde informasjon om de forskjellige brukerne.
Som interface mellom kunden og systemet vil vi bruke en touchscreen hvor kunden kan gi tilbakemelding om siste miks og hvor fornøyd den var med det. Her skal kunden også kunne velge hvor mye den vil handle for (kr). For at vi skal kunne lage en miks må vi ha en beregning som baserer seg på preferanser, tidligere valg og mengde.
Gruppen møttes i dronesonen og begynte å jobbe med mulighetene for å koble sammen flere motorer, slik at vi kan kjøre alle fire motorene samtidig. Dette ble gjort ved å modifisere koden fra forrige gang. Vi måtte også finne en annen strømkilde enn 9-volts batteriet som vi har brukt tidligere. Vi fikk litt problemer med to av motorene, ettersom de skal gå motsatt vei som de andre. Etter litt testing fant vi ut at vi hadde loddet to av motorene feil, ettersom de skal være counter-clockwise, så disse kjørte ikke ordentlig før vi fikk loddet på nytt. Etter riktig lodding og en god del koding fikk vi alle motorene til å kjøre samtidig. Deretter begynte vi å legge til ultra sonisk sensorer, slik at vi kunne eksperimentere med fartsøkning når den detekterer hindringer. Vi prøvde også å implementere for-løkker for å få en mykere start og stopp av motorene.
Gruppen møttes i dronesonen og fortsatte der vi slapp forrige uke. Vi fant fort ut at rammen vi allerede hadde laget til dronen var for liten, og Markus måtte derfor tegne og skjære ut en ny ramme. Resten av gruppa jobbet videre med å samkjøre ultra soniske sensorer sammen med motorene, og få disse til å kommunisere med hverandre basert på målinger som blir gjort. Vi monterte motorene på rammen for å se hvor mye de løfter, men vi fant ut at strømkilden vår ikke var tilstrekkelig for å drifte alle fire motorene på full styrke samtidig. Vi hadde noen problemer med å få alle til å kjøre samtidig, men dette ble fikset med litt koding.
Det jobbes iherdig med å få koden til å fungere som vi ønsker. Vi har testet ut forskjellige PID verdier og diverse andre verdier. Dette viste seg å være en utfordring. Men vi holder motet oppe (bl.a. med pepperkaker) og fortsetter å gruble på problemstillingen. Klikk på linken for å se hvordan roboten foreløpig ser ut i aksjon:
We have encountered problems with our raspberry pi noir camera, so we started experimenting with a new camera.
This camera is an ir camera, and not just a regular camera with the ir filter removed. This is a better alternative, since our images are not affected by the visible light in the room.
Our early experimentation looks very promising, with blood vessels clearly visible.
We will do more experimentation with our ir leds, to create a more even light, to improve image quality.
Hello world. We’ve now finally started assembling the car. The chassis is assembled, and wheels have been attached. The steering mechanism is attached as well.
we’ve also worked on programming the lidars and further developed our system. Its not long until we can start assembling the electronical components to the car and start testing.
see you soon,
Last tuesday (31.10.2017) we recieved the soundcard, and configured the Raspberry Pi to use it as default sound card. Today we ordered a mirror foil, expected to be delivered between 16th and 22th of November.
We made good progress on the software part of the smart mirror today. The face recognition module have been integrated with the main application and runs smoothly. Going forward we are going to integrate the registration module, as well as the speech recognition module.
We also decided to add a new option to the start-up screen, giving the user the ability to login as a guest. This means that there is no need for facial recogniton and no user data will be used, but instead some preset options will be used for both the news module, and the navigation module. The figure below shows a updated flowchart of the application.
The frame as it was shown, is printed with the Ultimaker at school (Figure 1).
Figure 1 First prototype 3D printed frame
The outer dimensions of the frame were almost good. The following dimensions are changed (Figure 2):
Figure 2 2D-drawing of the assembly of the frame
In according to these dimensions, it turned out that it would not be possible to put both servomotors inside the frame. Since the frame had the right dimensions in according to the arm, another system was required. We found out that it is possible to put the servomotors (1) outside the frame and use bolts and threads (2). Figure 3 shows the system for the 2-coordinate system.
Figure 3 The new 2-coordinate system
To make the pen move, there is another mechanical part required. For this movement system a 180° servomotor is used. The parts which connect the servomotor to the pen will be printed and assembled with bolts and nuts. Figure 4 and 5 show the system of the pen movement.
Figure 4 Mechanical part pen
Figure 5 Pen movement in system
The adjusted and additional parts are printed. Some parts needed to be adjust, because the holes were not big enough. The holes in the leading parts for the 2 coordinate system turned out smaller than supposed, causing fraction. The parts which are holding the pen were also smaller than supposed, so the holes in the parts had to be reamed. To lead the parts, thin copper axils are used. We use this material because it is very lightweight and strong enough to lead. Now that all the parts and materials are collected, we can start with saw the right dimensions of the thread and axils. After this the 2 coordinate system is assembled (figure 6). Also the mechanical part of the pen is assembled (figure 7).
Figure 6 Assembly 2-coordinate system
Figure 7 Assembly pen movement
While assembly the product, there were some complications. Three printed parts have to print again before the entire 2 coordinate system can be assembled.
All the parts are printed in the right way, so the system can be assembled again. Figure 8 shows the result of the assembled system.
Figure 8 Result of the assembled system
The video in the following link shows the working 2 coordinate system.
To connect the servomotor for the pen to the moving part in the middle, we had to adjust the servomotor. We had to polish the motor and move the wires to another place. Figure 9 shows the adjusted servomotor.
Figure 9 Adjusted servomotor
To connect the spindle, roll bearing and servomotor, we decided to print several parts. Figure 10 shows how these parts are assembled in a 3D-model and figure 11 shows an exploded view in a 2D drawing. Part 2 and 4, part 4 and 5 and part 3 and 6 (rollbearing and frame) are connected with glue. The servomotor and part 2 are squeezed together. In according to some tests it turned out it is strong enough when it is squeezed together. We chose this way to connect, because we cannot add glue to the servomotor, since it is a prototype and the servomotors will be used for other purposes.
Figure 10 3D-model of the connection between servomotor, roll bearing and thread
Figure 11 2D-drawing with an exploded view of the connection between servomotor, roll bearing and thread
Now the servomotors are only connected by the small pin. To make the servomotor connect more stable, we decided to print simple parts. We use these parts and double-sided tape so we can disassemble the motors after. Figure 12 shows a 3D-model of the connections and figure 13 shows a picture of the result.
Figure 12 Connection between servomotors and frame
Figure 13 Result of connection between servomotors and frame
After testing the servomotors, we concluded that this way of connection is good acceptable for the product.
The camera, LEDs and raspberry pi have to be connected to the frame. We decided to put the Raspberry Pi on top of the frame. According to this way, we have enough space to connect the wires. Figure 14 shows a 3D-model of the Raspberry Pi on the frame.
Figure 14 Raspberry Pi on top of the frame
The camera has to be connected underneath the frame (under the Raspberry Pi). Since we wnt to use the camera after this project, we cannot connect the camera with glue to the frame. We decided to print a part with holes in it which we can connect to the frame with double-sided tape. Figure 15 shows a 3D-model of the printed part and figure 16 shows a picture of the camera connected to the printed part.
Figure 15 3D-model of the printed part with nuts
Figure 16 Result of connection between camera and printed part
To make the camera work properly, we use 4 LEDs besides the camera. To connect the LEDs we decided to print another part. Figure 17 shows a 3D-model of the printed part together with the part for the camera.
Figure 17 Printed parts to connect camera and LEDs
The other part is printed and connected to the part with nuts, together with the LEDs. Figure 18 shows the result of the assembly.
Figure 18 Result of assembly camera and LEDs
To connect the device to the arm, we dicided to use Velcro. This is the easiest way to connect the device. Figure 19 shows the result of the device connected to the arm with Velcro.
Figure 19 device connected to the arm
We deciced that we are using another camera (IR camera), so we had to adjust a little bit to the frame. We are not using both the parts shown on figure 18 anymore, but only the black part. The new camera we are going to use is bigger than the other one, so it has to be connected on top of the frame in stead of underneath. According to this, we have to make a small hole in the frame for the camera. Figure 20 shows a 3D-model of the new method and figure 21 shows the result of the part connected to the frame. We are going to use souble sided tape to connect this part,
Figure 20 3D-model of the new method according to the camera
Figure 21 Result of the part to connect LEDs to the frame with double sided tape
The hole for the camera is drilled in the frame. While we were testing the device, it turned out that it is better to have more light to expose the arm. This is why we drilled more small holes in the frame, which makes us able to add more LEDs to the frame. Figure 22 and 23 shows the results of the drilled holes in the frame.
Figure 22 Result of the drilled holes in the frame
Figure 23 Result of the small holes in the frame for the LEDs
After testing, it turned out that the part to connect LEDs to frame with LEDs is not necessary anymore. Figure 24 shows a picture of the result without the connecting part.
Figure 24 Frame with extra LEDs and without connecting part
Following are the assembly drawings and all 2D drawings of the individual parts:
1. Assembly drawing robot needle
1.1 Main frame
1.3 Flat servo holder
1.4 Part reinforcement system
1.5 Connection motor – axle
1.6 Servo holder
1.10 Reinforcement top
1.12 Reinforcement sides
1.13 Leading part thread small
1.17 Leading part thread big
1.18 Leading part axle
2. Assemblydrawing pen
2.1 Connection pen short
2.2 Connection pen long
3. Assembly drawing connection thread – roll bearing – servo motor
3.1 Connection axle – rollbearing
Following document explains how the robotneedle is produced, step by step:
Hello world! today we had short meeting in the morning where we discussed how long until we could start putting pieces together. We found out that by next week we should be able to start putting the steering mechanism together. Computer had implemented a stepper motor for turning the car. We are not sure if we want to use a stepper or a servo but computer will look into getting code for servo as well to work. Hopefully our next post will be about assembly.
See you soon,