Model of cycling: closed loop with automatic adjustment of knee and pedal angles

I abused the Four Bar demo ( http://www.mathworks.de/help/toolbox/physmod/mech/gs/f5-6075.html) to create a cyclist-like geometry.

My problem is:

I don't get how I can change the variable segment/joint.

At the moment the "crank" adjusts its length according to the length and orientation of thigh and shank, but I need to adjust the angle of knee and pedal GIVEN the lengths of thigh, shank and crank!

http://imageshack.us/photo/my-images/52/blockdiagram20120210.png/

http://imageshack.us/photo/my-images/832/visualisation20120208.png/

Can anyone help me out? It's a blocker :(

br

Answer 1

[Guy Rouleau]

I am not sure I understand what you are looking for.

It looks like you did a very good job at computing the knee and hip torque based on the crank angle.

If you need to adjust the length of a body, you should split the inertia and mass of the body into 2 bodies and connect the 2 parts using a prismatic joint.

If you could include a schematic of what you are looking for, or a model showing what you tried, this could help.

Answer 2

[ZeuZ]

hello guy,

thanks a lot for your reply. sorry I didn't answer so far, had to earn some money...

and thanks for the comment on the torques :D my professor showed me mathlab two weeks ago and I sat in front of the montor like ... :O (????). he told me few things and I was really proud to have the leg make a full rotation (at least in the old version) by this angle-torque thing ;)

the "old" version: - no properties set for the crank (position of CS1 is 0 0 0) http://imageshack.us/photo/my-images/96/paramscrank.png/

- orientation of LOWER leg (shank) given in hard numbers in properties of the segment http://imageshack.us/photo/my-images/210/paramslowerlegold.png/

- orientation of UPPER leg (thigh) given in hard numbers in properties of the segment (similar to lower leg)

My Intention was to predefine the crank length of 0.1725 (vector [0.1725 0 0]) and upper leg length and lower leg length should be both 0.5 (or later take measures of a real athlete) and simulink should calculated the vectors automatically.

I have been working on another solution, manually calculating the (former hard numbered) vectors of thigh and shank and setting them by mathlab like this: http://imageshack.us/photo/my-images/526/paramslowerlegnew.png/ (sx = shank_x ; sy = shank_y ; msx = mass center shank_x ; msy = ...). I did the same for the upper leg (I'll skipp those screenies).

I put the files on my webspace for you... http://www.onpoint-media.com/Run_bike_2012_02_15.m http://www.onpoint-media.com/bike_opt_emg_2012_02_15a.mdl

you can adjust the range_crank and play time. the model is not capable of doing a full rotation yet, but I will work on that with an evolutionay algorithm. important thing was to be able to influence the crank length ;)

thanks a lot, again. I'll probably return severeal times during the next 6 month ;)

best regards, stefan

Answer 3

[ZeuZ]

aaaah, forgot...

open the model and go to file -> model properties -> callback -> initfcn to see the rest of the maths things

gotta find my clothes... swimming ;)