Pps with modulations dev #28
Conversation
…ndArmChains() - not tested
| @@ -11,6 +11,8 @@ | |||
| #define SKIN_THRES 7 // Threshold with which a contact is detected | |||
| #define RESP_GAIN_FOR_SKINGUI 100 //To amplify PPS activations from <0,1> to <0,100> | |||
| #define PPS_AGGREG_ACT_THRESHOLD 0.2 //Threshold for aggregated events per skin part | |||
| #define NR_ARM_JOINTS 7 | |||
| #define NR_TORSO_JOINTS 3 | |||
There was a problem hiding this comment.
Is it really needed to hard code a macro for the number of torso and arm joints?
Torso joints are already read from the encoder and stored in the jntsT variable. For the arm, you cannot read it from the encoders, but you can have a similar interface, i.e. a jntsAL variable that you initialize to 7 as a member of the class, instead of a #define. :)
There was a problem hiding this comment.
Matter of philosophy - macros make it IMHO more transparent.
There was a problem hiding this comment.
I know that, but since everything else is handled with another philosophy, it's better to stick to the existing one :)
There was a problem hiding this comment.
I prefer having subVector(0,NR_ARM_JOINTS -1) over subVector(0,6). If you object against macros, it can be made a variable with the same name (UPPER_CASE) and initialized in init(). It's slightly more work, but is the same. What is the problem with macro?
There was a problem hiding this comment.
The problem is that the other code uses jntsT jntsR, etc, so I would stick with that. Tha is I would use a variable (called, to be consistent, jntsAR jntsAL), initialized in the init as you said (this is what I suggested in the beginning).
jntsT is already initialized and used, so you can use that for the torso.
There was a problem hiding this comment.
|
Just tested everything - torso joints are taken into account now for remapping to taxel frames. Projections of taxel positions to camera frames are still operational too. |
|
Good! |
Torso joints are already read from the encoder and stored in the
jntsTvariable. For the arm, you cannot read it from the encoders, but you can have a similar interface, i.e. ajntsALvariable that you initialize to7as a member of the class, instead of a#define. :)subVector(0,NR_ARM_JOINTS -1)oversubVector(0,6). If you object against macros, it can be made a variable with the same name (UPPER_CASE) and initialized ininit(). It's slightly more work, but is the same. What is the problem with macro?jntsTjntsR, etc, so I would stick with that , i.e. with a variable (called, to be consistent,jntsARjntsAL), initialized in the init as you said (this is what I suggested in the beginning).jntsTis already initialized and used, so you can use that for the torso.eyeWrappersinto your new method. What do you think?I'm talking about lines
1102:1115and line1121Just need to test now.