I'm a master student trying to use MuJoCo to design an amphibious robot. To simulate the water, my current methodology is to adaptively change the viscosity, density, and wind speed of the simulation based on the position of the robot on a hfield amphibious terrain (see bottom right of video for values).
This somewhat works however is not ideal/accurate as fluid forces are being applied across the entire body whereas in real life in amphibious terrain only half the body would be in water.
Is it possible to apply various wind forces on various points of the hfield to mimic top half being air and bottom half being water. Any other suggestions or approaches would be much appreciated.
def controller(model, data):
global water_bounds
global rho_a
global beta_a
global rho_w
global beta_w
global valley_height
global fluid_vel
global wind_vel
global water_present
if water_present:
if valley_height <= water_bounds:
model.opt.viscosity = beta_w #viscosity of water
model.opt.density = rho_w #density of water
model.opt.wind = fluid_vel #speed of water (given it changes viscosity of entire system, wind will increase the speed of the entire denser/more viscous water flow as if its water flowing)
else:
model.opt.viscosity = beta_a #viscosity of air
model.opt.density = rho_a #density of air
model.opt.wind = wind_vel #no wind
else:
model.opt.viscosity = beta_a #viscosity of air
model.opt.density = rho_a #density of air
model.opt.wind = wind_vel #no wind
Hi,
I'm a master student trying to use MuJoCo to design an amphibious robot. To simulate the water, my current methodology is to adaptively change the viscosity, density, and wind speed of the simulation based on the position of the robot on a hfield amphibious terrain (see bottom right of video for values).
This somewhat works however is not ideal/accurate as fluid forces are being applied across the entire body whereas in real life in amphibious terrain only half the body would be in water.
https://github.com/google-deepmind/mujoco/assets/125491092/410d9807-f11b-404a-b461-976a430c01dd
Is it possible to apply various wind forces on various points of the hfield to mimic top half being air and bottom half being water. Any other suggestions or approaches would be much appreciated.