Steady-state visual evoked potential-based hierarchical obstacle avoidance brain-controlled wheelchair and control method

Abstract

This invention discloses a graded obstacle avoidance brain-controlled wheelchair and its control method based on steady-state visual evoked potentials (SVPs). By collecting EEG signals from the user's occipital lobe region and combining them with visual stimulation to induce steady-state visual evoked potentials, wheelchair movement commands are generated. Simultaneously, an environmental perception module acquires obstacle information in real time, and the wheelchair movement control module determines the obstacle avoidance status accordingly. Furthermore, the decoding process of the steady-state visual evoked potentials is constrained during the brain-controlled command generation stage, achieving coordinated control of obstacle avoidance warnings and brain-controlled commands. Through graded obstacle avoidance processing and corresponding feedback mechanisms, unsafe control commands are restricted at the generation stage, thereby improving the safety, stability, and applicability of the brain-controlled wheelchair in complex environments.

Description

Technical Field

[0001] This invention relates to the fields of brain-computer interface technology and intelligent rehabilitation assistive devices, specifically to a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials and its control method. Background Technology

[0002] Brain-computer interface (BCI) technology, by collecting and analyzing human brainwave signals, enables information interaction between the human brain and external devices, and has been gradually applied in the field of rehabilitation assistive devices. For users who have difficulty operating wheelchairs with their limbs due to neurological damage or motor dysfunction, brain-controlled wheelchairs based on BCIs can generate wheelchair movement commands in a non-contact manner, providing them with an alternative means of mobility control. Among existing brain-controlled wheelchair technologies, steady-state visual evoked potentials (SVPs) are widely used for wheelchair movement command recognition and control due to their high signal stability and relatively low training requirements. In addition, some existing solutions also incorporate environmental sensing devices to detect obstacles around the wheelchair and restrict wheelchair movement when potential risks are detected, thereby improving operational safety; however, existing brain-controlled wheelchairs based on SVPs still have certain shortcomings in practical applications:

[0003] 1. Existing obstacle avoidance mechanisms are mostly set up as independent safety control links at the wheelchair execution level. They usually intervene in wheelchair movement only after brain control commands have been generated, failing to restrict unsafe control intentions during the brain control command generation stage.

[0004] 2. In obstacle avoidance scenarios, users may continuously output brain control intentions pointing in the direction of obstacles. The system needs to repeatedly intercept or overwrite the generated control commands, which can easily lead to inconsistent control responses and increase the complexity of system decision-making.

[0005] 3. Steady-state visual evoked potential brain control is highly sensitive to user gaze behavior. When the operating environment is complex or obstacles suddenly appear, it is difficult for users to make quick corrections through brain control in a timely manner. Simply relying on the execution layer obstacle avoidance is not enough to reduce the generation of unsafe control commands from the source.

[0006] 4. Existing technologies lack a coordination mechanism that can effectively incorporate environmental obstacle avoidance information into the process of generating steady-state visual evoked potential brain control commands, making it difficult to ensure both safety and the continuity and stability of brain control interaction.

[0007] In view of this, the present invention proposes a graded obstacle avoidance brain-controlled wheelchair and control method based on steady-state visual evoked potentials. Summary of the Invention

[0008] The purpose of this invention is to provide a graded obstacle avoidance brain-controlled wheelchair and control method based on steady-state visual evoked potentials. This invention aims to solve the problems that existing brain-controlled wheelchairs based on steady-state visual evoked potentials lack a reasonable response mechanism to changes in the distance to surrounding obstacles in actual use, and lack an effective coordination method when user control commands and obstacle avoidance safety requirements coexist, which limits the safety and applicability of wheelchairs in complex environments.

[0009] In a first aspect, the present invention provides a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials, characterized in that it includes:

[0010] The EEG acquisition module is used to acquire EEG signals from the user's occipital lobe region in real time.

[0011] The visual stimulation module contains multiple stimulation units that flash at a preset frequency, and each stimulation unit corresponds to a wheelchair movement command.

[0012] The environmental perception module is used to obtain real-time information on the distance to obstacles around the wheelchair;

[0013] The EEG signal processing module is used to decode the EEG signals to generate wheelchair control commands;

[0014] The wheelchair motion control module is used to drive the wheelchair to move according to the wheelchair control commands and to feed back the obstacle avoidance status to the EEG signal processing module.

[0015] The EEG signal processing module includes a decoding control unit, which is used to dynamically constrain the judgment rules of steady-state visual evoked potential features based on different obstacle avoidance processing states fed back by the wheelchair motion control module, so as to restrict or prohibit the generation of wheelchair control commands related to the obstacle direction during the feature recognition stage of the EEG signal.

[0016] As a preferred technical solution of the first aspect of the present invention, the environmental perception module divides the distance to obstacles into at least three distance intervals, and different distance intervals correspond to different obstacle avoidance processing states.

[0017] As a preferred technical solution of the first aspect of the present invention, under different obstacle avoidance processing states, the decoding control unit restricts the generation process of brain control commands by increasing the recognition threshold of wheelchair movement commands related to the obstacle direction or by blocking the corresponding steady-state visual evoked potential feature determination channel.

[0018] As a preferred embodiment of the first aspect of the present invention, the visual stimulation module is linked with the obstacle avoidance processing state. When in the obstacle avoidance processing state where the generation of brain control commands is restricted, the stimulation unit corresponding to the restricted direction changes its flashing state or stops flashing, so as to reduce or suppress the generation of steady-state visual evoked potentials in the corresponding direction.

[0019] As a preferred embodiment of the first aspect of the present invention, the EEG signal processing module adopts a steady-state visual evoked potential recognition model based on correlation analysis, and the decoding control unit is used to dynamically adjust the reference signal weights or judgment parameters related to the corresponding stimulus frequencies according to the location of the obstacle.

[0020] As a preferred embodiment of the first aspect of the present invention, it further includes a human-machine feedback module, which outputs prompt information matching the corresponding obstacle avoidance level according to different obstacle avoidance processing states.

[0021] Secondly, the present invention provides a control method for a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials, for achieving the first aspect, comprising the following steps:

[0022] The system provides users with visual stimuli corresponding to different wheelchair movement commands and collects EEG signals from the user's occipital lobe region to induce steady-state visual evoked potential signals.

[0023] Simultaneously collect environmental information surrounding the wheelchair, including at least obstacle distance information;

[0024] The obstacle avoidance status of the wheelchair is determined based on the obstacle distance information.

[0025] Before generating wheelchair control commands, the obstacle avoidance processing state is introduced into the EEG signal decoding process to constrain the judgment rules of steady-state visual evoked potential characteristics.

[0026] Based on the aforementioned constrained decision rules, the EEG signals are decoded to generate wheelchair control commands with different executable states.

[0027] Based on the executability status of the wheelchair control commands, the wheelchair is controlled to perform the corresponding movement operations.

[0028] As a preferred embodiment of the second aspect of the present invention, the obstacle avoidance processing state is used to limit the set of wheelchair control commands that are allowed to be generated under the current environmental conditions, and the decoding of EEG signals and command generation are performed only within the scope of the set of commands.

[0029] As a preferred embodiment of the second aspect of the present invention, the generation of the wheelchair control command is based on the joint determination result of steady-state visual evoked potential characteristics and the obstacle avoidance processing state.

[0030] As a preferred technical solution of the second aspect of the present invention, when the obstacle avoidance processing state corresponds to the safety level that restricts the generation of wheelchair control commands, the display state of the visual stimulus corresponding to the restricted direction is adjusted to suppress the generation of steady-state visual evoked potentials in the corresponding direction from the source of induction; and during the EEG signal decoding process, the judgment parameters are adjusted to prevent the generation of movement commands pointing in the direction of the obstacle.

[0031] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0032] This invention incorporates the obstacle avoidance status obtained from environmental perception into the brain-controlled command generation stage during the control process of a brain-controlled wheelchair powered by steady-state visual evoked potentials (SVPs). By constraining the rules for judging SVP features, unsafe control commands related to obstacle directions are restricted or prohibited at the feature recognition stage, thus avoiding the control inconsistency caused by repeated interception or overwriting of generated commands during the command execution stage. Simultaneously, by dynamically adjusting the range of control commands that can be generated under different obstacle avoidance statuses, the probability of users continuously outputting unsafe brain-controlled intentions in complex environments is reduced, lowering the burden of brain-controlled interaction. Furthermore, by combining graded obstacle avoidance processing with corresponding human-machine feedback mechanisms, the wheelchair's operational safety status can be promptly and clearly communicated to the user. This achieves coordinated processing between brain-controlled intention generation and operational safety control at the system level, improving the safety, stability, and applicability of the brain-controlled wheelchair in complex usage environments. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0034] Figure 1 This is a schematic diagram of the functional module structure of the graded obstacle avoidance brain-controlled wheelchair of the present invention;

[0035] Figure 2 This is a schematic diagram of the control method for the graded obstacle avoidance brain-controlled wheelchair of the present invention. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings.

[0037] Throughout the accompanying drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions. The described embodiments are only a part of the embodiments of this application, not all of them. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0038] Example 1

[0039] Please see Figure 1 This embodiment provides a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials, including an EEG acquisition module, a visual stimulation module, an EEG signal processing module, a wheelchair motion control module, an environmental perception module, and a human-machine feedback module. Each module is connected through a data communication interface to form an overall system for performing brain-controlled wheelchair control and obstacle avoidance.

[0040] The EEG acquisition module is a non-invasive EEG acquisition device equipped with multiple electrodes to acquire EEG signals from the occipital lobe region of the user's head and transmit the acquired EEG signals to the EEG signal processing module.

[0041] To further explain, the non-invasive EEG acquisition module uses a dry electrode array, eliminating the need for conductive gel application, simplifying operation, and making it suitable for long-term wear. The electrode array is fitted to the user's occipital lobe and parieto-occipital junction to acquire steady-state visual evoked potential signals. The EEG acquisition module may have preset contact impedance ranges and sampling frequency ranges, and may include a preamplifier unit and an analog-to-digital converter unit to convert the acquired analog EEG signals into digital signals for subsequent processing.

[0042] For example, this area is the main generation area of ​​the SSVEP signal, which can improve the signal-to-noise ratio of the signal acquisition; the contact impedance of the dry electrode array is ≤100kΩ and the sampling frequency is ≥250Hz, ensuring that the detailed features of the SSVEP signal can be accurately acquired; the module has a built-in preamplifier (gain 1000-5000 times) and an analog-to-digital converter (sampling accuracy ≥16bit).

[0043] The visual stimulation module, located within the user's field of vision, includes multiple visual stimulation units operating at different flashing frequencies, each corresponding to a different wheelchair movement command. The EEG signal processing module is connected to the EEG acquisition module and is used to preprocess the acquired EEG signals and extract the steady-state visual evoked potential features corresponding to each visual stimulation unit.

[0044] Specifically, the visual stimulation unit may include multiple light-emitting units, distributed in a ring or other manner that facilitates user gaze within the user's field of vision, allowing the user to observe each visual stimulation unit without turning their head. Each visual stimulation unit may be driven with a different flicker frequency, selected from the frequency range recognizable by steady-state visual evoked potentials. The flicker frequencies of adjacent stimulation units may be set differently to reduce the possibility of frequency confusion.

[0045] In one example, different visual stimulation units can correspond to movement commands such as forward, backward, turning, or stopping of the wheelchair; the brightness of the visual stimulation units can be adjusted according to ambient lighting conditions. When obstacle avoidance conditions are triggered, the visual stimulation units can also provide prompts to the user by changing their display status.

[0046] For example, the visual stimulation unit includes at least four LED stimulation units, each corresponding to a core motion command of the wheelchair. The LEDs are driven to flash using a 6-40Hz square wave signal. This frequency range covers the sensitive frequency band of the SSVEP signal, and the frequency difference between adjacent stimulation units is ≥1Hz to avoid frequency confusion. Preferably, five core motion commands are set: forward (8Hz), backward (12Hz), left turn (16Hz), right turn (20Hz), and stop (24Hz). This frequency combination ensures the stability of the SSVEP signal while reducing user visual fatigue. The brightness of the LED stimulation units is adjustable (50-200cd / m²), which can adaptively adjust according to the ambient light intensity to improve visual clarity in different scenarios. Simultaneously, the LED stimulation units can respond to obstacle avoidance warning signals, strengthening the warning prompts by stopping flashing or changing color.

[0047] After recognizing the user's wheelchair control commands, the EEG signal processing module sends the corresponding control information to the wheelchair motion control module. The wheelchair motion control module, while receiving the control commands, processes them in conjunction with the detection results of the environmental perception module.

[0048] When the environmental perception module does not detect obstacle information that meets the obstacle avoidance conditions, the wheelchair motion control module drives the wheelchair to perform corresponding motion operations according to the control instructions; when the environmental perception module detects obstacle information that meets the obstacle avoidance conditions, the wheelchair motion control module adjusts or restricts the brain control instructions according to the preset control logic to avoid the wheelchair performing inappropriate motion operations under the current environmental conditions.

[0049] Specifically, the EEG signal processing module communicates with the EEG acquisition module wirelessly (Bluetooth BLE 5.0) or via wired connection. It incorporates an SSVEP signal enhancement algorithm, first removing power frequency interference through notch filtering (50Hz / 60Hz selectable), then preserving the effective frequency band of the SSVEP signal through bandpass filtering (1-70Hz), and finally using independent component analysis (ICA) to separate electrooculography artifacts, significantly improving signal purity. A decoding model based on canonical correlation analysis (CCA) is employed to quickly identify motion commands by calculating the correlation coefficient between the acquired EEG signal and the reference signal of the flashing frequency of each LED stimulation unit. When the EEG signal processing module receives obstacle avoidance status information from the wheelchair motion control module, it can restrict or mask the currently identified control commands to avoid continuing to output inappropriate motion commands even when obstacle avoidance conditions are met. The decoding latency is ≤200ms, meeting the requirements for real-time wheelchair control and early warning response.

[0050] The wheelchair motion control module presets at least two different distance ranges based on the obstacle distance information collected by the environmental perception module, and executes corresponding control processing logic for different distance ranges.

[0051] When the detected obstacle distance is within the first distance range, the wheelchair motion control module maintains the current motion state and outputs prompt information to the user through the human-machine feedback module; when the detected obstacle distance is within the second distance range, the wheelchair motion control module adjusts the wheelchair's motion parameters; when the detected obstacle distance is less than the preset safe distance, the wheelchair motion control module restricts or interrupts the wheelchair's drive signal.

[0052] Furthermore, the wheelchair motion control module can divide the obstacle distance information into multiple different distance intervals, each corresponding to a different control processing method. For example, when the obstacle distance is in a relatively far interval, the current motion state is maintained and a prompt message is output; when the obstacle distance is in the middle interval, the wheelchair's motion parameters are adjusted; when the obstacle distance is less than a preset safe distance, the wheelchair's drive signal is restricted or interrupted. The number of these distance intervals and the corresponding control methods can be set according to the wheelchair's operating status and usage scenario.

[0053] For example, the wheelchair motion control module includes a microcontroller (using an STM32H7 series chip), a motor drive unit, an attitude sensing unit, and a safety protection unit, wherein the safety protection unit is the core hardware carrier for realizing the obstacle avoidance and warning function.

[0054] (1) Hardware layout of safety protection unit: It integrates 4 ultrasonic ranging sensors, which are installed in the front left, middle and right directions and the rear center of the wheelchair respectively. The detection distance is 0.1-3m, realizing obstacle detection in all directions without blind spots, avoiding the risk of collision due to the omission of detection in a single direction; at the same time, it is equipped with an emergency braking button, which is installed on the side of the wheelchair armrest. It can be manually triggered by the accompanying person to cut off the motor power and realize emergency stop, as a supplementary protection means of automatic obstacle avoidance warning.

[0055] (2) Hierarchical obstacle avoidance warning logic: The microcontroller has three built-in warning thresholds and corresponding control strategies, specifically: The first warning threshold is set at 150-200cm (long distance warning). At this time, the obstacle is far away, and there is no need to adjust the wheelchair's movement state. Only a warning signal is output to prompt the user; The second warning threshold is set at 80-150cm (medium distance warning). At this time, the obstacle is close. The microcontroller controls the wheelchair to decelerate through the motor drive unit (deceleration ratio is 30%-50%), and outputs a warning signal at the same time; The third warning threshold is set at 50-80cm (close distance warning). At this time, the obstacle is extremely close. The microcontroller immediately triggers emergency braking, cuts off the motor drive signal, and makes the wheelchair stop quickly, while outputting the highest level warning signal at the same time; Each warning threshold can be adaptively adjusted according to the user's usage scenario to improve scenario adaptability.

[0056] (3) Attitude sensing unit collaboration: The MPU6050 six-axis sensor is used to collect the tilt angle and movement speed of the wheelchair in real time. On the one hand, it is used for anti-tipping adjustment, and on the other hand, the movement speed data is transmitted to the microcontroller to help optimize the obstacle avoidance warning strategy. For example, when the wheelchair speed is relatively fast, the warning threshold range of each level is automatically expanded to reserve a longer reaction and braking distance; when the speed is relatively slow, the warning threshold range can be appropriately reduced to improve passage flexibility.

[0057] An environmental perception module, installed on the wheelchair itself, is used to collect environmental information about the wheelchair's direction of travel and its surrounding area. The environmental information includes at least obstacle distance information related to the wheelchair's operational safety. The environmental perception module transmits the collected environmental information to the wheelchair motion control module in real time for obstacle avoidance judgment and control processing.

[0058] To further explain, the environmental perception module may include multiple distance detection units, which may be installed in a location that covers the area in front of, to the side of, or behind the wheelchair to improve the perception of the surrounding environment.

[0059] The human-machine feedback module is connected to the wheelchair motion control module and is used to output feedback information related to the wheelchair's motion status and obstacle avoidance handling to the user during wheelchair operation. The feedback information includes wheelchair motion status prompts and prompts corresponding to obstacle avoidance handling, so that the user can know the current operation status of the wheelchair.

[0060] Specifically, the human-machine feedback module may include an audio-visual prompt unit and a visual feedback unit, and output different forms of prompt information to the user according to the obstacle avoidance processing status. For example, under different obstacle avoidance levels, the prompt method is changed to provide feedback to the user on the obstacle status and the current processing result of the wheelchair, so that the user can know the operation status of the wheelchair.

[0061] For example, the human-machine feedback module includes an audio-visual prompt unit and a visual feedback unit, and designs a layered feedback mechanism for the obstacle avoidance warning function to ensure that users can quickly and clearly obtain warning information:

[0062] (1) Level 1 warning feedback: After receiving the Level 1 warning signal from the safety protection unit, the buzzer of the sound and light prompt unit emits a low-frequency short tone (frequency 1kHz, duration 0.3s, interval 2s) and the yellow indicator light flashes slowly (frequency 1Hz); the visual feedback unit (the display screen integrated into the visual stimulus coding module) displays the distance to the obstacle (accurate to 1cm), the direction of the obstacle, and the words "Level 1 warning", prompting the user to pay attention to the surrounding environment.

[0063] (2) Level 2 warning feedback: After receiving the level 2 warning signal, the buzzer of the sound and light prompt unit emits a short medium-frequency tone (frequency 2kHz, duration 0.3s, interval 1s) and the yellow indicator light flashes quickly (frequency 3Hz); the visual feedback unit displays the distance to the obstacle, the direction of the obstacle, the current speed of the wheelchair and the words "Level 2 warning - deceleration", so that the user knows that the wheelchair has entered the deceleration state and there is no need to issue an additional stop command.

[0064] (3) Level 3 warning feedback: After receiving the level 3 warning signal, the buzzer of the sound and light prompt unit emits a high-frequency long tone (frequency 3kHz, continuous sound until the warning is lifted) and the red indicator light flashes rapidly (frequency 5Hz); the visual feedback unit displays the distance to the obstacle, the direction of the obstacle and the words "Level 3 warning - braked"; at the same time, the LED stimulation unit of the current corresponding motion command stops flashing and turns red, and the warning effect is enhanced through multi-dimensional feedback to prevent the user from continuing to try motion commands because they do not notice the warning.

[0065] Example 2

[0066] This embodiment is based on the graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials described in Embodiment 1. Please refer to [link to Embodiment 1]. Figure 2 Furthermore, a control method for a brain-controlled wheelchair based on steady-state visual evoked potentials with a graded obstacle avoidance warning function is provided. This method focuses on real-time monitoring and response to obstacle avoidance warnings, and by incorporating the obstacle avoidance processing state into the brain control command generation process, it enables the safe operation of the brain-controlled wheelchair in complex environments. The method includes the following steps:

[0067] S1: The brain-controlled wheelchair system is activated and enters the initialization phase. The user wears the non-invasive EEG acquisition module. The working status of the EEG acquisition module is initialized to ensure that the EEG signal acquisition is available. At the same time, the visual stimulation module is activated so that multiple stimulation units generate stable visual stimulation signals according to a preset method. The environmental perception module is initialized, and at least two different obstacle avoidance processing states and their corresponding judgment conditions are set. The human-machine feedback module is self-checked to ensure that the feedback output function is normal.

[0068] Specifically, the microcontroller automatically calibrates the electrode contact state to ensure that the contact impedance of each electrode is ≤100kΩ; the visual stimulation encoding module is activated, and each LED stimulation unit flashes stably at a preset frequency; the safety protection unit is initialized, and three-level warning thresholds (such as level 1 180cm, level 2 120cm, and level 3 60cm) and corresponding control strategies (deceleration ratio 40%) are set; the feedback prompt module performs a self-check to ensure that the audio-visual and visual feedback is normal.

[0069] S2: The user gazes at the visual stimulus unit corresponding to the target wheelchair movement command according to the action requirements; when the user gazes at the stimulus unit, their visual cortex generates a steady-state visual evoked potential signal corresponding to the flashing characteristics of the stimulus unit; the EEG acquisition module collects the EEG signal in real time and transmits it to the EEG signal processing module; at the same time, the environmental perception module continuously collects information about the surrounding environment of the wheelchair, including at least obstacle distance information, and transmits the environmental information to the wheelchair movement control module in real time.

[0070] Specifically, when the target movement command is focused on the LED stimulation unit (e.g., focusing on an 8Hz flashing LED unit if forward movement is required), the visual cortex of the brain generates an SSVEP signal synchronized with the flashing frequency of the LED. The non-invasive EEG acquisition module acquires this signal in real time, and after pre-amplification and analog-to-digital conversion, it is transmitted to the EEG signal processing and decoding module. At the same time, the four ultrasonic ranging sensors of the safety protection unit acquire the distance data of surrounding obstacles in real time, and transmit the data to the microcontroller every 10ms.

[0071] S3: The wheelchair motion control module determines the current obstacle avoidance status of the wheelchair based on the obstacle distance information collected by the environmental perception module. The obstacle avoidance status is used to characterize the safety level of the wheelchair operation in the current environment. Before decoding the EEG signal, the EEG signal processing module receives the obstacle avoidance status and dynamically constrains the judgment rules of steady-state visual evoked potential characteristics based on the obstacle avoidance status to limit or prohibit the generation of wheelchair control commands related to the obstacle direction.

[0072] When obstacle avoidance restrictions are not triggered, the EEG signal processing module decodes the EEG signals and generates corresponding wheelchair control commands; when obstacle avoidance restrictions are triggered, the EEG signal processing module suppresses or blocks the generation process of control commands in the relevant directions.

[0073] Specifically, the received EEG signals are preprocessed (notch filtering, bandpass filtering, ICA artifact removal) to extract the time-frequency domain features of the SSVEP signal. The correlation coefficient between the preprocessed EEG signal and the reference signal of each LED stimulation frequency is calculated using the CCA algorithm. If the correlation coefficient corresponding to a certain frequency is greater than or equal to a preset threshold (e.g., 0.7), it is determined that the user has selected the movement command corresponding to that frequency, and the corresponding control signal is output. At the same time, the microcontroller compares the received obstacle distance data with the preset three-level warning threshold to determine whether a warning is triggered and the warning level, and outputs the warning signal of the corresponding level.

[0074] S4: The wheelchair motion control module receives the control command output by the EEG signal processing module and generates a wheelchair motion control signal based on the current obstacle avoidance status. When the obstacle avoidance status corresponds to the safety level that allows normal operation, the wheelchair motion control module drives the wheelchair to perform the corresponding movement according to the control command.

[0075] When the obstacle avoidance status corresponds to the safety level of restricted operation, the wheelchair motion control module adjusts the wheelchair motion parameters before executing the control command; when the obstacle avoidance status corresponds to the safety level of prohibited operation, the wheelchair motion control module suspends the execution of the user's motion control command and applies braking to the wheelchair; during wheelchair operation, the wheelchair motion control module can dynamically update the obstacle avoidance status according to the current operating status of the wheelchair.

[0076] This can be understood as follows: the wheelchair control module receives control signals and warning signals, and performs priority judgment: if no warning signal is received, the microcontroller controls the wheelchair to perform corresponding movements such as forward, backward, left turn, and right turn through the motor drive unit; if a first-level warning signal is received, the current movement state is maintained, and only feedback prompts are triggered; if a second-level warning signal is received, the microcontroller prioritizes controlling the motor drive unit to reduce the wheelchair speed (e.g., from the original speed of 2km / h to 1.2km / h), and then executes the user's movement commands; if a third-level warning signal is received, the microcontroller directly triggers emergency braking, cuts off the motor power, suspends the execution of the user's movement commands, and ensures that the wheelchair stops quickly; the posture sensing unit collects the wheelchair's tilt angle and movement speed in real time, and feeds the speed data back to the microcontroller. If the wheelchair speed is detected to exceed the preset safe speed (e.g., 3km / h), the warning threshold is automatically increased (first-level is adjusted to 200cm, second-level to 150cm, and third-level to 80cm).

[0077] S5: The human-machine feedback module outputs corresponding feedback information to the user based on the current obstacle avoidance status and wheelchair movement status; the feedback information is used to prompt the user about the current wheelchair operation status and safety level; when the obstacle avoidance status changes, the human-machine feedback module only outputs prompt information corresponding to the current obstacle avoidance status, so as to avoid the user making multiple control judgments during the brain control process.

[0078] Meanwhile, under obstacle avoidance processing conditions that restrict the generation of brain control commands, the visual stimulation module can synchronously adjust the display state of the stimulation unit corresponding to the restricted direction in order to reduce or suppress the generation of steady-state visual evoked potentials in the corresponding direction.

[0079] Specifically, the feedback prompt module provides corresponding information to the user based on control signals and warning signals: if no warning is triggered, the feedback command is successfully recognized and the wheelchair movement status is displayed; if a level one, two, or three warning is triggered, audio-visual and visual warning information is output according to the corresponding hierarchical feedback mechanism; if the obstacle is detected to have been removed (the ultrasonic ranging sensor does not detect the obstacle or the distance is >200cm), the microcontroller deactivates the warning state, the feedback prompt module resumes normal prompting, and the LED stimulation unit resumes normal flashing.

[0080] S6: Repeat steps S2 to S5 to achieve continuous operation of the brain-controlled wheelchair and full-process obstacle avoidance and early warning protection; when the user chooses to stop operation, terminate the generation of brain control commands and turn off the wheelchair motion control output to complete this use. After the wheelchair stops, turn off the system power.

[0081] Example 3

[0082] Based on the brain-controlled wheelchair system and its control method described in Embodiments 1 and 2, this embodiment, combined with typical complex traffic scenarios, provides an exemplary description of the operational effect of the control method of the present invention in actual use.

[0083] Turn on the system power by having the accompanying person press the main power switch on the side of the wheelchair, or by having the user trigger the switch with a slight head movement; a short "beep" sound will be heard and the green light on the feedback module will remain on, indicating that the system has started successfully; the system will automatically enter the self-test phase, which takes about 10 seconds, during which all the LED units of the visual stimulus encoding module will flash once in sequence.

[0084] Wearing and calibrating the non-invasive EEG acquisition module: The accompanying person assists the user in wearing the dry electrode array, ensuring that the electrodes are precisely attached to the occipital lobe and parieto-occipital junction (the upper part of the back of the head), and that the silicone base is completely attached to the head contour; the system automatically starts the electrode contact status calibration. If the green light on the feedback module remains constantly lit, it indicates that the electrode contact impedance is ≤100kΩ, which meets the requirements for SSVEP signal acquisition; if the green light flashes and is accompanied by a low-frequency beep, the accompanying person needs to slightly adjust the electrode position (no need to apply conductive gel) until the green light remains constantly lit, indicating that the calibration is successful.

[0085] Initialization complete: After the system self-test is completed, the buzzer will emit two "beep" sounds, and the feedback prompt module will keep the green light on. The five LED units of the visual stimulus encoding module will flash stably at preset frequencies, specifically: forward (left side, 8Hz), backward (right side, 12Hz), left turn (top, 16Hz), right turn (bottom, 20Hz), and stop (center, 24Hz). The 5-inch visual feedback screen will display "System ready, control can begin" and simultaneously display the current LED brightness level. At this point, the initialization is complete.

[0086] In practical applications, remember the LED command correspondence rules: The circularly distributed LED units are within the user's natural field of vision and can be clearly observed without turning their head. The command correspondence of each unit is as follows: Left LED (8Hz flashing) - controls the wheelchair to move forward, with a default speed of 2km / h; Right LED (12Hz flashing) - controls the wheelchair to move backward, with a speed limit of 1km / h (to ensure safe backward movement); Upper LED (16Hz flashing) - controls the wheelchair to turn left; Lower LED (20Hz flashing) - controls the wheelchair to turn right, with a turning angle of approximately 30° / second; Center LED (24Hz flashing) - controls the wheelchair to stop, suitable for regular pauses or emergencies.

[0087] Triggering basic motion commands includes: Triggering forward movement—look at the 8Hz LED unit on the left and maintain that gaze for about 0.2 seconds (system decoding delay ≤200ms), and the wheelchair will start moving forward; the visual feedback screen will simultaneously display "Forward speed: 2km / h", and the green light on the feedback module will remain constantly lit; Triggering turning—when switching from forward movement to left turn, move your gaze from the left LED unit to the 16Hz LED unit above and maintain that gaze for 0.2 seconds, and the wheelchair will automatically stop moving forward and start turning left; the visual feedback screen will update to display "Left turn angle: 30° / second"; Triggering backward movement—look at the 12Hz LED unit on the right and maintain that gaze for 0.2 seconds, and the wheelchair will slowly move backward; the visual feedback screen will display "Backward speed: 1km / h".

[0088] Emergency stop operations include: Brain-controlled stop (preferred): Focus on the 24Hz LED unit in the center, and the wheelchair will stop all movement within 0.2 seconds; Manual stop (backup plan): The accompanying person presses the DS-23 emergency brake button on the side of the wheelchair armrest, and the wheelchair will immediately stop by cutting off power; Stop feedback: Regardless of the stopping method, the visual feedback screen will display "Stopped" and the buzzer will emit a long beep to indicate to the user that the wheelchair has stopped.

[0089] The automatic obstacle avoidance function operates in conjunction with other functions (fully automatic, requiring no user intervention), including:

[0090] Level 1 Warning (obstacle distance 150-200cm): The feedback module flashes a yellow light slowly and emits a low-frequency short sound; the visual feedback screen displays "[XX direction] XXcm Level 1 Warning", and the wheelchair maintains its current movement.

[0091] Level 2 warning (obstacle distance 80-150cm): The feedback module flashes a yellow light quickly and emits a short mid-frequency sound; the visual feedback screen displays "[XX direction] XXcm Level 2 warning - deceleration", and the wheelchair automatically decelerates (example: from 2km / h to 1.2km / h).

[0092] Level 3 Warning (obstacle distance 50-80cm): The feedback module flashes a red light rapidly while emitting a high-frequency long tone; the visual feedback screen displays "[XX direction] XXcm Level 3 Warning - Braking," and the wheelchair immediately stops; the LED unit corresponding to the currently triggered motion command stops flashing and turns red. After the obstacle is removed, the system automatically cancels the warning, and the LED unit resumes normal flashing.

[0093] When obstacles in front of and to the sides of the wheelchair are detected to be within a safe distance, the EEG signal processing module generates corresponding motion control commands without obstacle avoidance restrictions, and the wheelchair maintains normal movement. As the wheelchair gradually enters a narrow area, when the distance to an obstacle in a certain direction enters the warning range, the wheelchair motion control module determines that the current obstacle avoidance status has changed and feeds this status information back to the EEG signal processing module. During the decoding phase, the EEG signal processing module constrains control commands pointing in the direction of the obstacle, while allowing turning commands unrelated to the direction of travel to continue to be generated.

[0094] In narrow spaces (such as corridors), the system automatically detects the distance between the two walls. When the distance is less than 100cm, a level 2 warning is triggered and the system slows down. The feedback screen displays "XXcm on the left and XXcm on the right". The user looks at the left / right turn LED unit to fine-tune the direction. The user keeps looking at the turn LED unit until the wheelchair is adjusted to a suitable position, and then switches to the forward command.

[0095] During this process, the wheelchair adjusts its motion parameters and, in conjunction with the human-machine feedback module, alerts the user to the current safe status, enabling the user to make minor directional adjustments and smoothly pass through narrow passages without additional thought or repeated input of commands. Once the distance to the obstacle is detected and returns to a safe range, the system automatically releases the corresponding obstacle avoidance status restriction, the EEG signal processing module resumes normal decoding of all motion commands, and the wheelchair continues to execute subsequent brain-controlled operations.

[0096] In environments with changing ambient light (indoor → outdoor), the system automatically adjusts the LED brightness (50-200 cd / m²) via an ambient light sensor, eliminating the need for manual operation. Under strong outdoor light, the feedback screen automatically brightens to ensure clear display. If the LED is still not clearly visible, accompanying personnel can assist in adjusting the angle of the LED module to ensure coverage of the user's field of vision.

[0097] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials, characterized in that, include: The EEG acquisition module is used to acquire EEG signals from the user's occipital lobe region in real time. The visual stimulation module contains multiple stimulation units that flash at a preset frequency, and each stimulation unit corresponds to a wheelchair movement command. The environmental perception module is used to obtain real-time information on the distance to obstacles around the wheelchair; The EEG signal processing module is used to decode the EEG signals to generate wheelchair control commands; The wheelchair motion control module is used to drive the wheelchair to move according to the wheelchair control commands and to feed back the obstacle avoidance status to the EEG signal processing module. The EEG signal processing module includes a decoding control unit, which is used to dynamically constrain the judgment rules of steady-state visual evoked potential features based on different obstacle avoidance processing states fed back by the wheelchair motion control module, so as to restrict or prohibit the generation of wheelchair control commands related to the obstacle direction during the feature recognition stage of the EEG signal.

2. The graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 1, characterized in that, The environmental perception module divides the distance to obstacles into at least three distance intervals, with each distance interval corresponding to a different obstacle avoidance state.

3. The graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 1, characterized in that, Under different obstacle avoidance conditions, the decoding control unit restricts the generation process of brain-controlled commands by increasing the recognition threshold of wheelchair movement commands related to the obstacle direction or by blocking the corresponding steady-state visual evoked potential feature determination channel.

4. The graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 1, characterized in that, The visual stimulation module is linked to the obstacle avoidance processing state. When it is in the obstacle avoidance processing state that restricts the generation of brain control commands, the stimulation unit corresponding to the restricted direction changes its flashing state or stops flashing in order to reduce or suppress the generation of steady-state visual evoked potentials in the corresponding direction.

5. The graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 1, characterized in that, The EEG signal processing module adopts a steady-state visual evoked potential recognition model based on correlation analysis. The decoding control unit is used to dynamically adjust the reference signal weights or judgment parameters related to the corresponding stimulus frequencies according to the obstacle's location.

6. The graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 1, characterized in that, It also includes a human-machine feedback module, which outputs prompt information that matches the corresponding obstacle avoidance level based on different obstacle avoidance processing states.

7. A control method for a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials, used to implement the graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials as described in any one of claims 1-6, characterized in that, Includes the following steps: The system provides users with visual stimuli corresponding to different wheelchair movement commands and collects EEG signals from the user's occipital lobe region to induce steady-state visual evoked potential signals. Simultaneously collect environmental information surrounding the wheelchair, including at least obstacle distance information; The obstacle avoidance status of the wheelchair is determined based on the obstacle distance information. Before generating wheelchair control commands, the obstacle avoidance processing state is introduced into the EEG signal decoding process to constrain the judgment rules of steady-state visual evoked potential characteristics. Based on the aforementioned constrained decision rules, the EEG signals are decoded to generate wheelchair control commands with different executable states. Based on the executability status of the wheelchair control commands, the wheelchair is controlled to perform the corresponding movement operations.

8. The control method for a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 7, characterized in that, The obstacle avoidance processing state is used to limit the set of wheelchair control commands that are allowed to be generated under the current environmental conditions, and the decoding of EEG signals and command generation are performed only within the scope of the set of commands.

9. The control method for a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 7, characterized in that, The wheelchair control commands are generated based on the joint determination of steady-state visual evoked potential characteristics and the obstacle avoidance processing state.

10. The control method for a graded obstacle avoidance brain-controlled wheelchair based on steady-state visual evoked potentials according to claim 7, characterized in that, When the obstacle avoidance processing state corresponds to the safety level that restricts the generation of wheelchair control commands, the display state of the visual stimulus corresponding to the restricted direction is adjusted to suppress the generation of steady-state visual evoked potentials in the corresponding direction from the source of induction; and during the EEG signal decoding process, the judgment parameters are adjusted to prevent the generation of movement commands pointing in the direction of the obstacle.

Images