Data Description
Within robotics, inverse dynamics algorithms are used to calculate the torques that a robot's motors must deliver to make the robot's end-point move in the way prescribed by its current task. More about [inverse dynamics][1].
[1]:
The data set consists of 48933 data points, collected at 100Hz from the actual robot performing various rhythmic and discrete movement tasks (this corresponds to 7.5 minutes of data collection).
Note that this version of the dataset contains both the train and test data from the original dataset, which includes many duplicates.
The task is to map from a 21-dimensional input space (7 joint positions, 7 joint velocities, 7 joint accelerations) to the corresponding 7 joint torques.
This version of the dataset includes only the training set of the original dataset, as the test dataset contains almost exclusively duplicates from the training data.
Attribute Description
1. *V[1-7]* - 7 joint positions
2. *V[8-14]* - 7 joint velocities
3. *V[15-21]* - 7 joint accelerations
4. *V[22-28]* - 7 joint torques, target variables, take one (*V22*) as target feature, ignore others as alternate target features