US 12,290,915 B2
Rehabilitation robot control apparatus and method thereof
Seung Kyu Nam, Seoul (KR); Ju Young Yoon, Suwon-si (KR); Tae Jun Lee, Suwon-si (KR); Beom Su Kim, Yongin-si (KR); Jae Seung Jeong, Daejeon (KR); Jae Hyun Kim, Daejeon (KR); Adedoyin Olumuyiwa Aderinwale, Daejeon (KR); Jun Ha Jung, Seongnam-si (KR); and Dong Hwa Jeong, Busan (KR)
Assigned to Hyundai Motor Company, Seoul (KR); Kia Corporation, Seoul (KR); and Korea Advanced Institute of Science and Technology, Daejeon (KR)
Filed by Hyundai Motor Company, Seoul (KR); Kia Corporation, Seoul (KR); and Korea Advanced Institute of Science and Technology, Daejeon (KR)
Filed on May 25, 2022, as Appl. No. 17/664,990.
Claims priority of application No. 10-2021-0088006 (KR), filed on Jul. 5, 2021.
Prior Publication US 2023/0001585 A1, Jan. 5, 2023
Int. Cl. A61H 1/02 (2006.01); A61H 37/00 (2006.01); B25J 9/16 (2006.01); B25J 11/00 (2006.01); B25J 13/08 (2006.01); G05B 13/02 (2006.01); G06F 3/01 (2006.01); G06N 3/008 (2023.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01)
CPC B25J 11/0005 (2013.01) [B25J 13/087 (2013.01); G05B 13/027 (2013.01); G06F 3/015 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A rehabilitation robot control apparatus comprising:
a processor; and
a memory with instructions stored thereon, wherein the instructions, when executed by the memory, enable to the apparatus to:
measure a brainwave signal of a user,
preprocess the measured brainwave signal by using a filter comprising a high pass filer, a band pass filter, and a notch filter, and by down sampling the brainwave signal,
preprocess the measured brainwave signal through conversion for time based on a predetermined window size and a degree to which a window is slid and overlapped,
receive information about whether a rehabilitation robot is operating or stopped,
extract a brainwave characteristic based on a band power for one or more brainwave measurement channels,
classify a motor intention of the user as a passive mode, an active mode, a rest mode or a motor imagery (MI) mode based on the brainwave signal,
classify the motor intention of the user while the rehabilitation robot is operating,
determine whether the user is in an active state in which the user has the motor intention or in a passive state in which the user does not have the motor intention while the rehabilitation robot is operating,
classify the motor intention of the user while the rehabilitation robot is stopped,
determine whether the user is in a motor imagery state in which the user has the motor intention or in a rest state in which the user does not have the motor intention while the rehabilitation robot is stopped,
convert the brainwave signal measured on a predetermined number of channels into an image sequence of a channel array status with a predetermined size and train a deep learning model based on a convolutional neural network (CNN) and a long short term memory (LSTM),
reflect the motor intention of the user in real time to control an operation or a stop of the rehabilitation robot, and
operate the rehabilitation robot, in response to a cumulative sum arriving at a predetermined threshold.