Sharpie pen style liquid markers where to buy
42 commentsTwitter account creator bot download
We tested the sensors and motors in the context of a semester thesis. Afterwards we designed and programmed an own robot that explores an unknown room and stores a map of this room in the Targa TGA image format. The semester thesis is divided into following subsections: Report on experiences and knowledge gained during work.
The first step on our list of tasks was to get a profound understanding of NXT's sensors and motors. Of primary interest was the ultrasonic sensor being the only one using digital communication with its own built-in micro-controller.
Despite its generally high accuracy, the sensor seems to have some weak points when measuring certain distances. The remaining sensors were tested as well, although not as extensively as the ultrasonic, the results are shown in the following subsections. In the end we took a closer look a the revised motors, now including rotation sensors. Following Parts of the NXT kit have been tested as part of the semester thesis: To get an idea of the ultrasonic sensor's accuracy and its rounding behavior we examined the sensors behavior towards objects within small distances.
The Object we used was a small cardboard box Distances smaller than 3 cm can not be measured. The biggest deviance of 2. The sensors mean deviance of 0. The second tests purpose was to gain knowledge of the sensors field of vision.
The same box as in the first test was used It was moved to different distances and angles relative to the ultrasonic sensor while noting the resulting distance readouts. The sensor was placed in horizontal as well as in vertical position, Figure 1. The results show that the ultrasonic sensor should always be placed in horizontal position, other positions decrease as well the field of vision as the sighting distance of the sensor. The sensor seems to be a bit 'blind' on the left eye, which can be explained by the fact that the left eye is actually the receiver of the ultrasonic wave while the right eye is the sender.
After these experiments on statical behavior of the ultrasonic sensor we moved on to some dynamic tests. The diagram in Figure 1. The data for this diagram was gained by writing a program in LEGO software based on LabView that stored the current ultrasonic value along with the current angle of one of the moving motors we used the TriBot model into a file on the NXT brick.
This file was then downloaded an processed. The dynamic test revealed two weaknesses of the ultrasonic sensor. The first issue is that it showed some areas where the sensor tends to measure cm instead of the actual distance. The second even more important issue is the critical area in between 25 cm and 50 cm where the sensor has a high probability of returning the wrong value of 48 cm.
The primary goal concerning the light sensor was to see to what amount it is able to distinguish different colors. The measurements where done in the sensor's reflected light mode. The results show that the sensor readout depends on the distance of the measured object, which makes it difficult to generally assign certain colors to sensor values. The touch sensor has been examined towards the force needed to close the touch circuit.
Adding weights to the vertical positioned sensor showed that at 34 grams the appears as touched. So in general a force of 0. The motor has been tested towards its linearity between power and rotation speed. A very rudimentary perl script that is able to download a file from the brick has been elaborated and is available in the Thesis Results section.
The perl script uses the Win As part of the thesis we had to plan and conduct a project. The goal of the project was to build a fancy robot that uses the capabilities of the new "LEGO Mindstorms NXT" generation to the limits of the included sensors and motors.
We decided to go for a map building robot that can explore any given room, hence the name Explorer. The project is divided into following subsections: The walls of the unknown room should be aligned in more or less regular angle to each other.
As concluded by J. Map building Motion planning Collision avoidance Localization These tasks have been implemented separately and in the given order but for the localization function omitted because of lack of time.
The program was elaborated through foollowing stages: Explorer 1, Explorer 2 and Explorer 3. A robot deemed fit for the needed requirements was constructed. The robot then travels forward 5 cm and repeats the first step and so on. Motion planning Collision avoidance Start positioning The main idea here was to explore the room in a clockwise direction by always following the wall on the left side.
Once the robot reaches the position it originated from the program stops. The result we achieved with our Explorer project was satisfying. It got worse tough, the lower the batteries got. The thesis and its Powerpoint summary are written in German and are available for download.
Further below are the source code and binary of the explorer project, followed by the perl script to communicate with the NXT brick.