A Method and Device for Control of a Mobility Device

Abstract

A system for control of a mobility device comprising a controller for analyzing data from at least one sensor on the mobility device, wherein the data is used to determine the gait of user. The gait data is then used to provide motion command to an electric motor on the mobility device.

Description

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH

Not applicable.

BACKGROUND OF THE INVENTION

The invention relates to a mobility device. More specifically, the invention relates to a control system and method of controlling a mobility device having an electric motor that is worn on the feet of a user to provide mobility assistance.

Commuters and other travelers often have to walk the final leg of their trip, regardless of whether they traveled by car, bus, train, or other means. Depending on the distance, the time needed to complete this final leg of the journey can comprise a significant amount of the total duration of the trip. While bikes or scooters can be used, they are bulky and require skill and a minimum level of fitness to operate. Powered systems, such as moving walkways, suffer from a lack of mobility. Other mobility solutions suffer the same drawbacks or lack the ability to adapt to a particular user. Therefore, it would be advantageous to develop a control system for a mobility device that does not require any special skills or user training and can adapt to the individual needs of a particular user.

BRIEF SUMMARY

According to embodiments of the present invention is system and method of controlling a mobility device, wherein the mobility device is worn on each foot of a user. A sensor obtains data about the gait of a user and transmits the data to a processor. The processor analyzes the gait of a user and then uses the gait data to develop motion commands for each mobility device. The mobility device may comprise a motor, gearing, and wheels. When worn on the feet of a user, the mobility devices allow a user to walk at an increased rate of speed for a given cadence and stride length, as compared to their speed without the mobility devices. Further, the control system adapts to a user so no learning or other control inputs are required by the user.

BRIEF SUMMARY OF THE SEVERAL VIEWS OF THE DRAWINGS

FIG. 1 depicts a mobility device with an embedded controller, according to one embodiment.

FIG. 2 is a block diagram of a control system according to one embodiment.

FIG. 3 shows the steps of the method of control, utilizing the controller depicted in FIG. 2.

DETAILED DESCRIPTION

As shown in FIG. 1, a mobility device 100, according to one embodiment, comprises a plurality of wheels 101, with at least one of the wheels 101 connected to an electric motor 102. Further shown in FIG. 1 is an onboard controller 111 and an optional remote controller 112. During typical use, a user will wear two mobility devices 100, one on each foot. The mobility device 100 enables a pedestrian to walk faster than a normal walking pace by adding torque to the wheels 101 of the mobility device 100 worn on the foot in contact with the ground. In this manner, the user experiences an effect similar to that of walking on a moving walkway. More specifically, the control system 110 of the present invention enables a user to maintain a normal walking motion by adapting the control of the motor 102 to the movements of the user. As will be discussed in greater detail, the speed at which the wheels 101 spin, through a torque applied by the motor 102, is controlled in part by an analysis of the user's gait.

FIG. 2 depicts the components of the onboard controller 111, which comprises at least one inertial measurement unit 113, a processor 114, a motor driver 115, and a wireless communication module 116. Two onboard controllers 111 are shown in FIG. 2 since each mobility device (i.e. one for each foot of the user) will house an onboard controller 111. In an alternative embodiment, the control system 110 may also include a remote controller 112, which is capable of sending commands to each of the onboard controllers 111. In this particular embodiment, both the left and right mobility devices 100 receive command speeds from the remote controller 112, which can be in the form of a hand-held controller, a computer, or a mobile phone, and actuate the mobility devices at the specified command speeds.

The control system 110 is used to collect data and analyze the gait of a user. For example, the onboard processor 114 reads gait dynamic data, comprising acceleration, gyroscopic data, and quaternion data of each mobility device 100 from the inertial measurement unit 113. In one embodiment, both onboard controllers 111 send the gait dynamic data to the remote controller 112 and, in return, receive a motion command from the remote controller 112. The motion command comprises, for example, acceleration to a set speed, braking, deceleration to a set speed, and holding at a constant speed. In alternative embodiments, additional data can be included in the motion command. Alternatively, the motion command may be generated by the onboard controllers 111. Upon receiving the motion command, the onboard processor 114 along with the motor driver 115 converts the motion command into a motor driving signal and drives the motor system 102, thereby affecting the speed of the wheels 101. In one embodiment, the motor driver 115 receives a speed command and drives the motor 102 at the command speed via a feedback loop control.

The flow diagram shown in FIG. 3 depicts the method of gait-based motion control, according to one embodiment, comprising the steps of receiving gait dynamic data 301, determining the user gait 302, and determining the motion command 303.

In step 301, the remote controller receives gait dynamic data from both onboard controllers 111. The gait dynamic data includes data collected from the inertial measurement unit 113. Next, at step 302, the user gait is determined in step 302 by testing data through the machine learning model. More specifically, the remote controller receives the gait data and predicts the user's gait based on a trained model. In one embodiment, step 302 comprises feeding the gait dynamic data from a prior step into the beginning of a fixed size data buffer. When new data is received, the oldest data is discarded from the data buffer. The size of the buffer can be sufficiently large to cover at least one full gait cycle of the gait dynamic data. The data buffer is then fed into a pre-trained machine learning model to determine the user gait. According to one example embodiment, the machine learning model is a support vector machine. However, alternative machine learning models can be used. The machine learning model is trained based on the user performing various gaits on mobility devices 100 and signaling her current gait to the control system 110 via an input on the remote controller 112. At step 303, the motion command is generated based on the determined gait.

However, in optional step 304, the remote controller 112 checks if any user input has been registered. The user input can be in various forms such as pressing a button or moving the remote controller 112 in a certain trajectory. For example, the user input may press a button indicating that the user wants forward motion. Thus, the forward motion command received from the user can override the motion command provided by the controller 112 based on the machine learning model. After checking for a user input at step 304, a motion command is generated and sent by the remote controller 112 to both onboard controllers 111. However, if the user input is received from step 304, the final motion command is replaced with the user input before being sent to the onboard controllers 111.

In an alternative embodiment, each onboard controller 111 generates a motion command and sends the motion command signal to the other controller 111 for cross-validation in step 305. The motion command may include acceleration to a set speed, braking, deceleration to a set speed, and holding at a constant speed. Upon validating the motion command, the processor 114 along with the motor driver 115 convert the motion command into a motor driving signal and drive the motor system. Stated differently, in step 305, cross validation compares the motion commands generated for each of the two mobility devices 100. For example, the motor driver 115 will only command motor speed when both commands are similar and will brake when the speed commands are inconsistent.

While the disclosure has been described in detail and with reference to specific embodiments thereof, it will be apparent to one skilled in the art that various changes and modification can be made therein without departing from the spirit and scope of the embodiments. Thus, it is intended that the present disclosure cover the modifications and variations of this disclosure provided they come within the scope of the appended claims and their equivalents.

Claims

1. A method of controlling a mobility device having an electric motor, the system comprising: receiving gait data from at least one inertial measurement unit,determining the gait of a user based on the gait data and a pre-configured machine learning model; andgenerating a motion command using the determined gait.

2. The method of claim 1, wherein the gait data is selected from the group consisting of acceleration, gyroscopic data, and quaternion data.

3. The method of claim 1, wherein determining the gait of a user comprises testing the gait of a user through a machine learning model.

4. The method of claim 3, further comprising: training the machine learning model by having a user perform various gaits on the mobility device and signaling the various gaits to a control system.

5. The method of claim 1, further comprising cross validating the motion command between two mobility devices worn by a user.

6. The method of claim 5, wherein the step of cross validating the motion command comprises: converting the motion command into a motor driving signal if the motion command of a first mobility device is similar to a motion command of a second mobility device.

7. The method of claim 5, wherein the step of cross validating the motion command comprises: converting the motion command into a braking signal if the motion command of a first mobility device is not similar to a motion command of a second mobility device.

8. The method of claim 1, further comprising: checking for user input from a remote controller, andoverriding the motion command based on the user input.