test_gelslim

1 Basic Features

Demonstrate GelSlim tactile sensor simulation.

2 Implementation Process

2.1 Initialize the Environment

env = RFUniverseBaseEnv(ext_attr=[GelSlimAttr])

2.2 Demonstrate GelSlim Tactile Sensor Simulation

gelslim = env.InstanceObject(name="GelSlim", attr_type=GelSlimAttr)
gelslim.SetTransform(position=[0, 0, 0])
target = env.InstanceObject(name="GelSlimTarget", attr_type=attr.RigidbodyAttr)
target.SetTransform(position=[0, 0.03, 0], rotation=[90, 0, 0])
env.SetViewTransform(position=[-0.1, 0.03, 0.], rotation=[0, 90, 0])

InstanceObject to instantiate the GelSlim sensor.
SetTransform to set the sensor and object at appropriate positions and orientations.
SetViewTransform to set the viewing angle at an appropriate position and orientation.

for i in range(50):
    env.step()
    target.AddForce([0, -1, 0])

AddForce to apply the given force to a rigid body.

gelslim.GetData()
env.step()
image = np.frombuffer(gelslim.data["light"], dtype=np.uint8)
image = cv2.imdecode(image, cv2.IMREAD_COLOR)
cv2.imshow("light", image)
cv2.waitKey(0)
image = np.frombuffer(gelslim.data["depth"], dtype=np.uint8)
image = cv2.imdecode(image, cv2.IMREAD_GRAYSCALE)
cv2.imshow("depth", image)
cv2.waitKey(0)

Call GetData on the GelSlim sensor object to retrieve the sensor data, then use the cv2 module to display it.

gelslim.BlurGel()
gelslim.GetData()
env.step()
image = np.frombuffer(gelslim.data["light"], dtype=np.uint8)
image = cv2.imdecode(image, cv2.IMREAD_COLOR)
cv2.imshow("light", image)
cv2.waitKey(0)
image = np.frombuffer(gelslim.data["depth"], dtype=np.uint8)
image = cv2.imdecode(image, cv2.IMREAD_GRAYSCALE)
cv2.imshow("depth", image)
cv2.waitKey(0)
env.Pend()
env.close()

Calling BlurGel on the GelSlim sensor object can blur the gel mesh, simulating smooth deformation.