The limitation of the ultrasonic sensor in the underground drilling robot avoids obstacle is analyzed

and the scheme solved is put forward.To emphasize that

temperature and humidity compensation to ultrasonic wave sensor is carried on.Try to use Elman feedback neural network to approach function.Elman network latent layer adopts "Tansig" activate function

output layer activate function with"Pureline"

this guarantees that once the network has enough layers

it can approach wanton function with wanton precision.Experiment proves that ultrasonic ranging’s measurement precision can raise two orders of magnitude after carying on the temperature and humidity compensation.The working efficiency of obstacle avoidance in the system is improved greatly

the ability of drilling robot obstacle avoidance is improved.