Intelligent walking aid

Abstract

A smart walking aid device has a walker frame having a plurality of substantially vertical sides and intersecting front portions. The wheels are located at both ends of the bottom edges of the sides. A soft robot sensing handle extends in a C shape along the upper edges of the sides and the front portion. The sensing handle has a plurality of contiguous inflation chambers, each of which includes a pressure sensor for generating a pressure signal characterizing a pressure within the chamber. A microcontroller unit receives a pressure signal from a pressure sensor of the handle chamber and determines a status of at least one of the device and a user of the device based on the pressure signal. A stability augmentation mechanism is driven by the microcontroller to stabilize the walker in response to the determined state of at least one of the walker and the user.

Description

Technical Field

[0001] The utility model relates to an intelligent walking aid, which is used to support the elderly or disabled people when walking, and in particular to a walking aid that uses artificial intelligence and multiple sensor data fusion to provide more reliable, convenient and functional walking assistance. Background Art

[0002] The ability of older people and people with disabilities to safely care for themselves is receiving more attention due to the rapid increase in the elderly and disabled population. The number of older people will exceed 2 billion by 2050. World Health Organization (2018). World Health Statistics 2018: Monitoring health to achieve the Sustainable Development Goals, SDG 613 [dataset] Wikipedia Contributors (2020). Azimuth - Wikipedia, the free encyclopedia. [Online; accessed 614 on 8 February 2020]. Of all the daily activities, walking is essential for people with disabilities and older adults to achieve an active and independent quality of life. Mobility support not only prevents a slow decline in a person's overall functioning, but also reduces healthcare costs by delaying premature hospitalization.

[0003] A variety of walkers have been developed to address this problem. However, most existing walking assistance products are designed using traditional methods with only a few rigid joints. As a result, the level of assistance they can provide is very limited. See the following articles: Ferrari, Francesco et al., “Human-robot interaction analysis of an intelligent walker for the elderly: ACANTO interactive guidance system,” International Journal of Social Robotics 12.2 (2020): 479-492; Werner, Christian et al., “Evaluation of an intelligent walker for sit-to-stand transition assistance for elderly people with motor impairments,” International Journal of Gerontology and Gerontology 20.4 (2020): 312-316; and Valadão, Carlo et al., “A novel controller for an intelligent walker based on human-robot formation,” Sensors 16.7 (2016): 1116. However, these walkers all face problems, such as not being able to provide adequate and safe support or being inconvenient in various modes of human-robot interaction. According to Luz, Clare, Tamara Bush, and Xiaoxi Shen, “Are Canes Different Than Walkers? Uselessness and Fall Injuries,” Gerontologists 57.2 (2017): 211-218, approximately 33% of people over 60 have experienced at least one fall, despite the assistance of a walker.

[0004] There are only a handful of smart and interactive walking assistance devices, and their prices are beyond the reach of most families. These include U.S. Patent No. 10,850,401, entitled "Soft Robotic Actuator and Processing Method Thereof," which has a mechanism for actively expanding a soft chamber to generate motion, and a soft sensing handle with a fixed amount of fluid, and passively changes shape due to grip pressure to give feedback and provide protection. U.S. Patent No. 10,767,024, entitled "Soft Actuator and Preparation Method," discloses a method for preparing a soft actuator. U.S. Patent No. 10,780,591, entitled "Soft Robotic Actuator and Gripper," discloses a soft sensor. U.S. Patent No. 10,597,275, entitled "Soft Robotic Actuator for Positioning, Packaging and Assembly," discloses a mechanism for positioning, packaging and assembling a walker. According to U.S. Pat. No. 10,576,643, entitled "Sensor for Soft Robots and Soft Actuators," a strain limiting layer is added to the soft actuator and a sensor is embedded in the strain limiting layer. In U.S. Pat. No. 10,465,723, entitled "Soft Robot Actuator," there is at least one strain limiting wall, while the soft sensing handle does not have a strain limiting wall. Finally, in U.S. Pat. No. 10,385,886, entitled "Soft Actuator and Soft Actuation Device," linear motion can be generated by deflation, where the soft sensing handle passively changes shape to measure the user's state and absorbs force to protect the user.

[0005] In the previously designed intelligent walker disclosed in the article "Intelligent Robotic Walker with Intelligent Close-Range Interaction Capability for Elderly Travel Safety", Frontiers in Neurorobotics 14 (2020), by Zhao, Xiaoyang et al. ("Zhao Design"), the device has multiple basic functions. They are: 1) a novel mechanical mechanism with omnidirectional motion design and reliable stability, which can prevent users from falling and provide sufficient fall protection; 2) an artificial intelligence-based front-following model, which can analyze the movement of the walker by observing the user's gait through an infrared thermal imager, and then drive the walker to follow the user in front; 3) sound source localization (SSL) based on reinforcement learning, which can detect the horizontal orientation of the sound source, allowing the user to recall the walker from a distant position; 4) a soft robotic handle, which can be used as an interface to predict the user's intention and detect the emergency of falling, which can also activate the fall protection of the walker.

[0006] An improved smart elderly / disabled walking aid with a user-friendly and effective interface that can greatly improve the user experience while increasing reliability, stability, and convenience at an affordable price would be advantageous in the art. Utility Model Content

[0007] The utility model aims to solve the instability and high cost of the existing system through an intelligent walking aid for the elderly and disabled.

[0008] The utility model relates to an intelligent walker, which comprises: a walker frame, which has a plurality of substantially vertical sides and an intersecting front portion, with a plurality of wheels located at both ends of the bottom edges of the sides; a soft robotic sensor handle, which extends in a C-shape along the plurality of sides and the upper edge of the front portion, the sensor handle having a plurality of adjacent inflatable chambers, each of the inflatable chambers comprising a pressure sensor for generating a pressure signal representing the pressure in the chamber; a microcontroller unit, which receives a pressure signal from the pressure sensor of the chamber of the handle, and a stable software module running on the microcontroller determines the state of at least one of the intelligent walker and the user of the intelligent walker based on the pressure signal; and also comprises an inertial measurement unit sensor, which measures the orientation of the intelligent walker and sends data to the microcontroller, and when the walker is operating normally When the system is in a protection state, the microcontroller ignores data from the inertial measurement unit sensor, and when the system is in a protection state and the inertial measurement unit sensor detects that the smart walker is tilted, the microcontroller will activate a stabilization mechanism to stabilize the smart walker and provide stable support to the user; and multiple stabilization mechanisms are arranged around the bottom of the frame and driven by a stabilization software module of the microcontroller to stabilize the walker in response to the determined state of at least one of the walker and the user; and it also includes multiple microphones facing the top of the walker and the bottom of the walker, and a sound source localization software module of the microcontroller, which detects the direction of the person relative to the walker based on the triangulation of the human voice received by the microphone when summoning the walker.

[0009] That is to say, the intelligent walker includes three components: a soft robotic sensor handle, a microcontroller unit and a stabilization mechanism. The soft sensor handle is placed on the upper part of the intelligent walker, the microcontroller unit is placed on the front middle part of the walker, and the stabilization mechanism is placed on the bottom of the walker. The working principle of the utility model is as follows: the user places both hands on the soft sensor handle, and the pressure sensors embedded in the multiple chambers of the soft handle measure the pressure in each chamber. Based on the pressure data, the microcontroller determines what state the user and the walker are in and whether they are suitable for that posture. The intelligent walker will respond to any inappropriate posture. For example, if the intelligent walker tilts and the tilt angle exceeds a preset value, the microcontroller will send a control signal to trigger the stabilization mechanism, thereby stabilizing the intelligent walker and protecting the user from falling.

[0010] The soft robotic handle of the walker provides comfortable and stable support for the user and can protect the user when they fall. The soft robotic handle has a series of pressure sensors that detect the user's state and intention. The stabilization mechanism is combined with the soft handle to provide strong and stable support to the user. When the soft handle detects that the user is about to fall, the microcontroller immediately triggers the stabilization mechanism and activates the support points on the side and front of the smart walker so that the walker is fully fixed and can provide support for the user.

[0011] The use of soft materials for the handle allows the user to interact with the smart walker in a safer and more comfortable way. The stabilization mechanism follows a modular design (e.g., three units are spaced around the walker and activated separately). In addition, the size of each unit is minimized so that the entire mechanism does not increase the overall size of the walker. However, the stabilization mechanism is strong and operates fast enough to provide support to the walker. By combining this handle with a new leg stabilization design, the walker is able to avoid tipping over and provide enhanced stability.

[0012] The utility model may also include an infrared thermal imager facing the user's calf and a front follow model (FFM) unit that analyzes a series of images from the camera using a long short-term memory (LSTM) recurrent neural network designed to collect walking data. This system is able to predict the user's walking gait with higher prediction accuracy and frequency than the prior art, as well as predict the user's intention (e.g., whether the user is about to turn) together with sensor information from the soft robot handle. Using data from the camera and handle sensors, i.e., data fusion, the user's walking assistance is improved. As a result, the powered wheels located at the lower rear of the walker can guide it to the front of the user at the expected speed and in the expected direction. In addition, by changing the walking information collection and calculation process, the utility model achieves a higher reaction frequency, allowing the walker to perform better with higher gait prediction accuracy and efficiency.

[0013] In addition, a new Sound Source Localization (SSL) unit is provided, which includes multiple microphones towards the top of the walker and the bottom of the walker to calculate the direction of the sound source near the walker and the vertical displacement of the sound source. Thereby, the user can give a voice recall if the walker is nearby. The walker can then move towards the user by using its motor-driven wheels so that the user can place on the walker. In addition, the SSL unit can determine the vertical displacement of the user, i.e. whether the user is in a standing posture, a sitting posture or a prone posture (indicating a user who has fallen). The upgraded soft robotic handle has more pressure sensors, which can collect more user and walker data and can provide fast and robust pressure feedback. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1is a rear left top perspective view of an intelligent walker or a walking aid according to an embodiment of the utility model, wherein an inset diagram shows an enlarged view of two parallel springs connecting the front wheels and main components of the walker;

[0015] Figure 2 A front follower system used in conjunction with the walking aid of the present invention is shown;

[0016] Figure 3A is a schematic diagram of three essential components of a walking aid according to an embodiment of the utility model, Figure 3B The soft sensing handle, its cross section, and the effect of pressure on it are shown;

[0017] Figure 4A It is a schematic diagram of three working states controlled by the microcontroller of the walking aid of the utility model and the transition between the states. Figure 4B is a schematic diagram showing four user states and the transitions therebetween during the use of the walking aid of the utility model;

[0018] Figure 5 is a block diagram of a control unit of the present utility model; and

[0019] Fig. 6A It is a three-dimensional diagram of the stabilization mechanism of the utility model. Figure 6B It is a side view of the connecting rod of the stabilization mechanism. DETAILED DESCRIPTION

[0020] Figure 1 An embodiment of a walker of the present invention is shown, along with the locations of the main sensors and mechanical components. Typically, the walker frame has two sides and a front structure, forming a space in the middle for the user to stand. Small wheels 9 are located at the lower rear of the frame, and large wheels 11 are located at the lower front of the frame. The rear wheels 9 can be driven by a motor 10 located above each rear wheel 9. These can be powered by a battery carried by the walker. A set of two parallel springs 4 (inset) connect the front wheels 11 to the main components of the walker on both sides. These springs facilitate travel over bumpy surfaces.

[0021] When the walker is started, for example, when the walker is about to fall over, three stabilizing mechanical units 8 provide a stable support for the walker. Each of these modules is placed on the left and right sides of the walker, and one module is placed under the front frame.

[0022] A segmented soft robotic handle 3, which may be made of silicone, is arranged along the top edge of the frame. Each segment of the handle 3 has its own inflatable chamber. The segments are bonded together one by one to form a complete C-shaped handle. Pressure sensors are located in each chamber to detect where along the handle 3 the user is gripping the handle and how hard they are gripping.

[0023] The middle platform 5 of the walker extends at the front between the two sides. This platform 5 contains a microcontroller, which can be a processor or a group of processors, a battery and some sensors, such as a GPS sensor, an inertial measurement unit (IMU) sensor and some infrared distance detection sensors. Here, its signal is provided to the microcontroller, which combines the signal with the radar signal to improve map building and obstacle avoidance. The infrared thermal imager 2 is arranged in the middle position of the middle platform at the front of the walker. It faces the direction in which the user stands. A heart rate sensor 6 can also be arranged on the body of the walker. A light detection and ranging sensor 7 (for example, a ranging radar sensor) is further located on the body and can scan the surrounding environment in a 2D plane. That is, it is capable of scanning the surrounding environment of the smart walker in a two-dimensional plane and sending radar information to the microcontroller for map building and obstacle avoidance so as to navigate the walker by selectively driving the wheels of the walker.

[0024] The frame is equipped with six microphones 1, which may be of the same type, to detect sound signals from the user. Four of the microphones are located at each corner of the top of the frame, and two microphones are located at the front or back at the bottom of the frame. The sound signals collected by the microphones are analyzed in the microcontroller as part of a sound source localization (SSL) software module that can locate the direction from which the user's voice comes as part of the walker summoning system.

[0025] The walker performs the following in the base scene:

[0026] a) Assume that the walker is initially far from the user and the user may have fallen or it is inconvenient for the user to move to the walker. According to one embodiment of the present invention, the walker is smart enough to respond to a voice call from the user so that it moves to a position near the user.

[0027] b) When the walker is nearby, assuming the user is standing, the user grabs it and starts operating it. If the user has fallen, the user can lean on the walker or otherwise use the walker to recover from the fall.

[0028] c) When the user walks with the walker, the walker can provide full and intelligent walking assistance by moving in front of the user with the user's estimated walking gait, and can detect the user's intention and move in the direction indicated by the intention. In addition, if the walker senses that the user is about to have a problem, such as an indication that the walker is starting to fall, the walker can deploy stabilization arms to learn and protect the user and the walker.

[0029] d) At the same time, due to the soft nature of the arms where the user contacts the walker, the user experience is comfortable and smart.

[0030] Compared to the prior art, the present invention includes a number of improvements, one of which is the ability of the walker to move towards the user. This is achieved through a sound source localization (SSL) system based on reinforcement learning, which analyzes sound signals collected by multiple microphones. When the user summons the walker, the microphones will receive acoustic signals from the user's voice at different times. By analyzing the differences in reception times at different microphones, the system can calculate the direction of the sound source, i.e., from which direction the sound comes, by triangulation. That is, the sound source localization software module detects the direction of the person relative to the walker when the walker is summoned based on the triangulation of the person's voice received by the microphones. Based on this detection, the rear wheels of the walker can be driven to move the walker towards the sound source, i.e., the user. The sound source localization system can be a software module running in the microcontroller of the walker or in a separate processor.

[0031] In the existing Zhao design, the walker only has four microphones located at different positions in the same horizontal plane. As a result, the walker cannot detect the height of the sound source. The characteristic of the utility model is that by adding at least two microphones at different horizontal planes, the walker can detect the height of the sound source. Figure 1 As shown, four microphones 1 are arranged at the four corners of the upper surface of the walker, and two more microphones are arranged on the main body below the other microphones. The additional microphones give the walker the ability to detect whether the user is lying on the ground. In addition, in conjunction with map construction, obstacle avoidance, and radar sensors 7 and Figure 1 With the addition of other sensor-dependent navigation capabilities, the system is able to design a safe and fast path for the walker to move toward the user after the SSL calculates the sound source orientation.

[0032] The utility model also includes a novel recovery procedure and user status monitoring. After the walker is close to the user, the user can Figure 1The soft robot handle 3 in the walker is characterized in that it has a larger detection area and more fall protection than the Zhao design. Similarly, the soft robot handle is introduced into the intelligent design so that it can detect the user's fatigue, condition, intention and emergency when the user operates the walker. By monitoring the average maximum pressure value of the user acting on the handle chamber within a preset time period, when the pressure value is getting larger and larger, the soft robot handle 3 can distinguish that the user is becoming more and more tired. That is, by monitoring the average maximum pressure value of the user acting on the chamber of the soft robot handle within a preset time period, abnormal high pressure and / or sharp pressure changes can be detected as an emergency of the user. When the user walks, the pressure value and distribution applied to the handle are different from when the user is standing still. Walking intention can also be detected by comparing the pressure values ​​and changes acting on different areas of the handle. For example, when the user turns left, the pressure on the left side will be greater than the pressure on the right side because the user is more inclined to the left side. Through a large detection range and a high detection frequency, the handle can detect abnormally large pressure and sharp pressure changes caused by emergency situations such as falls. When the soft handles mounted on the lower part of the walker are grasped, this indicates that the user may have fallen and is trying to get up by himself by grasping these handles, which is an example of an emergency situation.

[0033] The utility model provides intelligent walking assistance. Specifically, when the user is ready to operate the walker, the front follower system starts to work. It can detect and analyze the user's gait information for walker control and can drive the walker to follow the user's gait when it is located in front of the user. 24×32 pixel infrared thermal imager 2 ( Figure 1 ) is arranged in the front middle position of the middle platform 5 of the walker to capture the thermal image of the user's lower limbs. Instead of using an optical camera, the infrared thermal imager protects the user's privacy while capturing the main features of the user's gait. With the help of gait information, the front following model utilizes a recurrent neural network to achieve fast response and high precision of the control module. In the existing Zhao design, there is a front following model that uses a deep neural network (DNN) module instead of a recurrent neural network to analyze data from infrared thermal gait imaging, and the position of the legs is corrected by a radar sensor. The radar in the Zhao design has a low scanning frequency, which brings a low prediction frequency to the module. In addition, Zhao's stabilization mechanism consists only of a braking system, which cannot maintain the stability of the walker when pushed by a large horizontal force.

[0034] In the present utility model, Figure 2The system of the front-following model shown in processes the infrared image data (multiple consecutive 24×32 gait images) to achieve a higher prediction frequency. The frequency limit is only limited by the frequency of the infrared camera, which is faster than the infrared camera used in Zhao's design.

[0035] In order to obtain higher prediction accuracy, the module of the present invention creatively uses a convolutional neural network ("CNN") to extract image features, and then uses a long short-term memory ("LSTM") recurrent neural network to analyze several consecutive feature frames. The LSTM results are output to the user motion prediction module to obtain the front-following ability. The module used in the present invention has a network training accuracy of over 96%, while the network in Zhao's design only reaches about 90%. This feature enables the walker to provide more adequate walking assistance. It does not require any sensors directly connected to the user. The processing of the front-following ability can be performed by a software module executed on the microprocessor 20 of the walker. It can also be executed on a stand-alone processor.

[0036] According to the utility model, a data fusion function is included to further improve walking assistance. Specifically, the utility model is characterized by obtaining some data fusion from different sensors. When the user walks, the pressure acting on the soft robot handle changes. By adding data from the soft robot handle to the network, the front follower model can better synchronize the prediction of gait motion and user intention. For example, when the user turns left, the center of gravity will be more biased to the left and the pressure on the left side will increase.

[0037] The present invention also provides a more comfortable and safe walking interaction through the combination of soft robotic handles and stabilization mechanisms: this is achieved in part by providing adequate walking support in different environments, including indoor and outdoor situations. Therefore, the structure has a shock absorption system, which is added to the two front wheels. Each wheel has two horizontal springs to maintain the stability of the walker. The shock absorption system not only improves the user's experience in a bumpy environment, but also reduces the noise signal of detecting pressure changes from the soft skin handle. That is, two horizontal springs connect each front wheel to the walker frame to maintain the stability of the walker in a bumpy environment and reduce the noise signal of the walker during travel.

[0038] In order to make the user aware of his / her condition, a new feature of the present invention is that it includes a GPS sensor and a heart rate sensor. The GPS sensor can tell the user the location of the walker when it is outdoors. That is, the GPS sensor provides the user with information about the geographical location of the walker. The heart rate sensor can detect the user's heart rate simply by the user placing a finger on it for a few seconds. These new sensors allow the user to clearly monitor his / her condition while operating the walker. This allows the walker and the user to detect more data and stay away from some potential dangers, such as heart attack or getting lost.

[0039] like Figure 3A As shown, the soft sensor handle 3 is a combination of different sections, each of which has a separate inflatable chamber. The shape of the soft sensor handle and the number of sections inside a handle are customized so that it can be placed in different positions of the smart walker to serve as an interface and protection for the user. For ease of understanding, a semicircular cross-section is used to illustrate its working principle, as shown in FIG. Figure 3B The soft handle body can be made of low-cost and skin-friendly silicone rubber, and it is prepared so that an internal sealed chamber is created and air is trapped inside. When the user's hand is holding on to the soft handle, the holding pressure will cause the soft handle to deform ( Figure 3B The pressure inside each chamber will change accordingly. The internal pressure of each chamber is measured by a pressure sensor located inside the chamber. Based on the pressure change distribution, the system can determine what state the user is in, and thus determine the working state of the smart walker. The air trapped in the chamber provides fast and robust pressure feedback.

[0040] The smart walker has three stable states in total. Figure 4A . Each of these three states can be changed to another state, except the locked protection state which cannot be changed to the warning state. In most cases, the user will gently grasp the soft handle with pressure within the normal range. In this case, the system will be in the unlocked normal state and all functions can be performed normally. From the normal state, if the pressure increases significantly but is still within the preset range, and the rate of change is also within the acceptable range, the system will change from the normal state to the warning state, and some control capabilities will be disabled.

[0041] There are three conditions for the system to enter the protection state: a) one or more pressure values ​​have exceeded the acceptable range; b) one or more pressure sensors have a significant sudden increase; c) the number of detected pressure changes exceeds the preset value. If one or more of the above conditions are met, the system will enter the protection state, regardless of what state the system is initially in. When the system is in the protection state, all movement functions fail, that is, the walker stops moving, but the system is able to provide support and protection for the user.

[0042] To release the self-protection state, the user needs to gently place their hand on the soft sensing handle. This creates an unlocked pressure data pattern that will be detected. When this pattern is detected, the system will transition from the protection state to the normal state. The system can also transition from the warning state to the normal state, triggered by the pressure falling into the normal range. In different states, the smart walker will have different performance.

[0043] The soft sensor handle can also be used as a user status monitor. There are four user states (such as Figure 4B As shown), it can be classified by the pressure data obtained from the soft sensing handle. If the soft sensing handle has measured a series of constant and continuous pressure data, the user is standing and holding the soft handle with both hands. If the pressure data conforms to a sine wave distribution, the user is walking with the smart walker. If one side of the soft sensing handle measures a greater pressure than the other side within a certain period of time, the user is turning. If in any of the above three states, the average pressure has increased significantly within a certain period of time, the smart walker will consider that the user is tired and take corresponding measures.

[0044] like Figure 3A As shown, the pressure signal from the soft sensor handle 3 is transmitted to the microcontroller 20, which determines what action the walker will perform, including extending the arm or link 30 of the stabilization mechanism 8. The control unit used in the present invention mainly has three necessary components, the microcontroller 20, the inertial measurement unit sensor 22 and the pressure sensor matrix 24, as shown in FIG. Figure 5 As shown. The pressure sensor matrix measures the pressure in the air chamber in the soft handle and sends the pressure signal to the microcontroller 20. The inertial measurement unit sensor 22 measures the orientation of the intelligent walker based on the pressure signal and sends the data to the microcontroller. The microcontroller will control the intelligent walker and the stabilization mechanism 8 based on the acquired data.

[0045] The functions of the stabilization control unit are as follows: a) The pressure data from the soft handle is measured by the pressure sensor matrix 24 and processed in the microcontroller 20. The microcontroller will then decide what state the system is in. b) The orientation of the intelligent walker will be measured by the inertial measurement unit sensor 22 and transmitted to the microcontroller. c) When the system is in the normal state or the warning state, the data from the inertial measurement unit sensor will be ignored by the microcontroller. d) When the system is in the protection state and the inertial measurement unit sensor detects that the intelligent walker is tilted, the microcontroller will activate the stabilization mechanism 8 to stabilize the intelligent walker and provide a stable support for the user.

[0046] exist Fig. 6A and Figure 6BThe stabilization mechanism 8 shown in the figure follows a modular design. In the current embodiment of the utility model, the system is divided into three modules, which can be placed in any convenient position on the intelligent walker. In the current embodiment, the module 8 is installed at the lower left, lower right and lower front of the walker. The mechanism is designed so that the size is minimized so as not to increase the overall size of the walker, and is strong and fast enough to provide support for the walker.

[0047] The components of the stabilization mechanism 8 are: a servo motor 35, a shaft seat 36, a main stabilization link 37, an extension shaft 38, a push rod 39, a stepper motor 40, a shaft seat 41, a push rod 42, a second link 43, a linear bearing 44, a link 45, a spring 46, a shaft 47 and a first link 48. The working principle of this mechanism is as follows: in a normal state or a warning state, the stabilization mechanism 8 is in a default position in which the main stabilization link 37 is parallel to the ground. When the system enters a protection state, the stabilization mechanism 8 is triggered. The servo motor 35 turns the push rod 42 upward so that it no longer blocks the first link 48. Next, the spring 46 pushes the first link 48 toward the shaft seat 41, and the second link 43 will extend downward, and the main stabilization link 37 will extend downward from a parallel position toward the ground to provide support for the intelligent walker. This entire process is completed within 0.1 seconds to ensure that the intelligent walker can be quickly stabilized to provide support and protection for the user. When the system is released from the protection state, the stepper motor 40 will drive the push rod 39 to push the main stabilization link 37 back to its default parallel position. Figure 6B When the main stabilization link 37 is in its default position, the servo motor 35 will rotate the push rod 42 downward to lock the mechanism, so that the two motors are not loaded in the default position.

[0048] Although the utility model is explained in conjunction with certain embodiments, it should be understood that various modifications thereof will become apparent to those skilled in the art after reading this specification. Therefore, it should be understood that the utility model disclosed herein is intended to cover these modifications falling within the scope of the appended claims.

Claims

1. An intelligent walking aid, comprising: a walker frame having a plurality of generally vertical sides and an intersecting front portion, a plurality of wheels being located at either end of a bottom edge of the sides; a soft robotic sensing handle extending in a C-shape along a plurality of sides and an upper edge of the front portion, the sensing handle having a plurality of adjacent inflatable chambers, each of the inflatable chambers including a pressure sensor for generating a pressure signal representative of the pressure within the chamber; a microcontroller unit that receives a pressure signal from a pressure sensor of a chamber of the handle, a stable software module running on the microcontroller determining a state of at least one of the intelligent walker and a user of the intelligent walker based on the pressure signal; Also included is an inertial measurement unit sensor, which measures the orientation of the intelligent walking aid and sends data to the microcontroller, when the walking aid is operating normally, the microcontroller ignores the data from the inertial measurement unit sensor, and when the system is in a protection state and the inertial measurement unit sensor detects that the intelligent walking aid is tilted, the microcontroller activates a stabilization mechanism to stabilize the intelligent walking aid and provide a stable support to the user; and a plurality of stabilization mechanisms disposed about a bottom portion of the frame and driven by a stabilization software module of the microcontroller to stabilize the walking aid in response to a determined state of at least one of the walking aid and a user; as well as It is characterized in that it also includes multiple microphones facing the top and bottom of the walker, and a sound source localization software module of the microcontroller, which detects the direction of the person relative to the walker based on the triangulation of the human voice received by the microphone when summoning the walker.

2. The intelligent walking aid according to claim 1, characterized in that , wherein the sound source localization software module determines whether the person summoning the walker is in a prone position based on triangulation between sounds received from a microphone toward the top of the walker and sounds received from a microphone toward the bottom of the walker.

3. The intelligent walking aid according to claim 1, characterized in that , also includes a GPS sensor that provides the user with information about the geographic location of the walker.

4. The intelligent walking aid according to claim 1, characterized in that , also includes a light detection and ranging sensor that can scan the surrounding environment of the smart walker in a two-dimensional plane and send radar information to the microcontroller for map construction and obstacle avoidance in order to navigate the walker by selectively driving the wheels of the walker.

5. The intelligent walking aid according to claim 4, characterized in that , also includes an infrared distance detection sensor, whose signal is provided to the microcontroller, and the microcontroller combines the signal with the radar signal to improve map building and obstacle avoidance.

6. The intelligent walking aid according to claim 1, characterized in that , wherein the intelligent walking aid has three stable states, namely, an unlocked normal state, a locked protection state, and a warning state, and The intelligent walker enters the protection state in the following cases: a) one or more pressure values ​​from the soft robot handle have exceeded the acceptable range; b) one or more pressure sensors have a significant sudden increase; and c) the number of pressure changes detected within a time period exceeds a preset value; Wherein when the intelligent walking aid is in the protection state, all movement functions are disabled, and the system can provide support and protection for the user; wherein when the user gently places his hand on the soft robot sensor handle to generate an unlocking pressure data distribution that causes the intelligent walker to switch from the protection state to the normal state, the intelligent walker is released from the protection state; and When the pressure drops to a normal range, the intelligent walking aid switches from a warning state to a normal state.

7. The intelligent walking aid according to claim 1, characterized in that , also includes a heart rate sensor mounted thereon, which is capable of detecting the user's heart rate when the user places his finger on it for a few seconds.

8. The intelligent walking aid according to claim 1, characterized in that , wherein the microcontroller is a single processor or multiple processors.

9. The intelligent walking aid according to claim 1, characterized in that , wherein by monitoring the average maximum pressure value applied by the user to the chamber of the soft machine handle within a preset time period, abnormally high pressure and / or rapid pressure changes can be detected as an emergency situation of the user.

10. The intelligent walking aid according to claim 1, characterized in that , also includes two horizontal springs that connect each front wheel to the walker frame to maintain the stability of the walker in a bumpy environment and reduce the noise signal of the walker during travel.

11. The intelligent walking aid according to claim 1, characterized in that , where the stabilization mechanism includes: a main stabilization link, wherein when the stabilization mechanism is in a default position and the walker is in a normal unlocked state, the main stabilization link is parallel to the ground; When the stabilization mechanism is triggered because the walker is about to enter a protection state, the servo motor rotates the push rod upward so that it releases the first connecting rod; a spring, which pushes the first connecting rod toward the axle seat; The second link is hinged to the first link and extends downward, and pushes the main stabilizing link downward from a parallel position toward the ground to provide support for the intelligent walker.

12. The intelligent walking aid according to claim 11, characterized in that , wherein the stabilization mechanism further comprises: a stepper motor, wherein when the intelligent walking aid is released from the protection state, the stepper motor drives the second push rod to push the main stabilization link back to its default parallel position; and When the primary stabilization link is in its default position, the servo motor rotates the first push rod downward to lock the mechanism, so that the two motors are not loaded in the default position.

Images