This is pure Python code that can read a *.kinarm exam file.
Image Display Spring
This isn’t really a task per se, it’s really just a demonstration of how to dynamically manipulate an image and use forces. When you start the task an image of a spring will be displayed between the hands. As you try to move your hands apart or together you will feel a spring trying to hold you at one distance. The task will work on a Kinarm Exoskeleton or EP robot.
The task was built for Dexterit-E 3.9 and MATLAB 2015a.
Note: This is a much improved version of the task. This has been cleaned up substantially since the first version.
Input File Example – Drawing Hand Paths
This example makes use of the Input File block. This is a block that was introduced in Dexterit-E 3.9. The Input File block allows that task programmer to specify that a data file should be passed to the task when it runs. The operator needs to specify the file that will be passed.
In this example, a previously recorded exam is processed to output a set of X,Y hand positions. Those positions are passed to the Robot Computer so that the task can draw the hand positions with a set of Path VCodes.
The task is compiled for MATLAB 2015a SP1 and Dexterit-E 3.9
Show Target for 1 Frame on the Subject Display
This task is designed to show a target for a single frame on the subject display. The algorithm should work properly for any refresh rate on the subject display. The model is extensively commented with the details of the algorithm.
The task is compiled for MATLAB 2015a SP1 and Dexterit-E 3.7.
Visualization of Equations of Motion Parameters
This is a demo on how to access some of the data about the robots (link lengths, angles, positions of hand, elbow, and shoulder) in real time during a task. The task will draw the data on the subject display in real time to show what the link lengths, angles, positions of hand, elbow, and shoulder are. It is a real time visualization of Fig 7-1 Equations of Motion Parameters from the Dexterit-E 3.7 User Guide.
As well, the task demonstrates the use of moving targets, targets with changing labels, and the path target type, new in Dexterit-E 3.7.
The task is compiled in MATLAB 2013a and Dexterit-E 3.7 and will not work on earlier Dexterit-E versions.
Hand Trail
This is a task that will demonstrate the use of path targets, new in Dexterit-E 3.7. The task will draw a trail that will follow hand movement.
Compiled in MATLAB 2013a and Dexterit-E 3.7 and will not work on earlier Dexterit-E versions.
Coherent Dots Discrimination Task
This demonstration task is intended to show the use of the repeat target type, new in Dexterit-E 3.7, and is based on the ideas presented in:
Britten, Kenneth H., Michael N. Shadlen, William T. Newsome, and J. Anthony Movshon. “The analysis of visual motion: a comparison of neuronal and psychophysical performance.” Journal of Neuroscience 12, no. 12 (1992): 4745-4765.
Pilly, Praveen K., and Aaron R. Seitz. “What a difference a parameter makes: A psychophysical comparison of random dot motion algorithms.” Vision research 49, no. 13 (2009): 1599-1612.
This task was compiled in MATLAB 2013a and Dexterit-E 3.7 and will not work on earlier Dexterit-E versions.
Saving Data at 2 kHz
This task will demonstrate how to save data at 2 kHz in a Kinarm task (which saves data at 1 kHz). The task will generate a 2 kHz signal that counts from 0 and splits the task into 2 signals (the unmodified signal and another signal that is delayed by 1/2 ms). By saving both signals, we sample at 2 kHz.
Also included is a MATLAB script read_2kHz.m that will recombine the two separate signals into the original 2 kHz signal generated in the model.
Compiled in MATLAB 2013a and Dexterit-E 3.6.4.
KINVASION!
This is a task written by Ethan Heming, a Ph.D. canditate in Steve Scott’s lab. Kinarm would like to thank Ethan for allowing us to post his task! This is easily the most sophisticated task program I have seen. This game really stretches what can be done with Simulink, Stateflow and indeed Dexterit-E.
This task requires a bilateral Kinarm End-Point robot with force-torque sensors.
The object of this game is to destroy as many enemies as possible on each level. You control a ship represented by a green bar between your hands. The thickness of this line shows your ship’s health. Any enemies or enemy shots that hit this bar reduce your health and the game ends when ship has no health left. If you get hit there will be a haptic response (i.e. you will feel it)! At each handle you have a “gun” that fire perpendicular to the green health bar. They are fired by ’tilting’ robot handle (applying a pitch or roll torque). The farther apart your hands are, the faster your guns recharge, but the larger a target you present to enemies.
At the start of each level you can buy upgrades to your guns and shields by moving the cursor to the desired selection and torquing on the handle. There are also addition options such as auto-cannon or bombs. During a level you can catch coins to earn money to buy things. You also earn coins by killing enemy ships.
Some extra special features are:
If you have an End-Point robot with force plates then stomping on the plate will release a bomb if you have purchased that upgrade.
I have compiled this task for Dexterit-E 3.8 and MATLAB 2015a.
Have fun!
Visuomotor Rotation Task
This task is provided to demonstrate how to create a visuomotor rotation task.
This task also provides an example of a “persistent” variable in embedded MATLAB (see the Calculate_Desired_Rotation function with /Visual_Rotation/Triggered Subsystem/).
Task was built using MATLAB 2013a and Dex 3.5