Adaptive artificial microsaccade event camera binocular system and control method thereof

Abstract

This invention discloses an adaptive artificial micro-eye-tracking event camera binocular system and its control method. The adaptive artificial micro-eye-tracking event camera binocular system includes a left event camera, a right event camera, and a mounting plate. The left and right event cameras have built-in inertial measurement units (IMUs) for acquiring angular velocity data. A motor is mounted on the mounting plate, and the motor's output is connected to a transmission pinion. Rotatable left and right prism fixing gears are located on the front side of the left and right event cameras on the mounting plate. Left and right wedge prisms are mounted on the left and right prism fixing gears. A photoelectric switch assembly is mounted on the mounting plate. This invention aims to meet the needs of mobile robots for three-dimensional spatial perception in navigation, obstacle avoidance, and other tasks, and can adaptively adjust according to the camera's motion state to balance event triggering efficiency in low-dynamic scenes, error compensation in high-dynamic scenes, and system energy consumption.

Description

Technical Field

[0001] This invention relates to the field of computer vision technology, specifically to an adaptive artificial micro-eye movement event camera binocular system and its control method. Background Technology

[0002] Inspired by the micro-eye-tracking mechanism of the human visual system, research has proposed the Artificial Microsaccade-enhanced Event Camera (AMI-EV) technology. This technology places a rotatable wedge prism in front of the event camera's aperture. By rotating the wedge prism, the deflection direction of the incident light rays is actively changed, causing the incident light to form an approximately circular motion trajectory on the image plane. This allows for the active triggering of event responses at scene edges even when the camera is stationary or moving slowly, reducing information loss caused by edges being parallel to the camera's motion direction. Combined with a motion compensation algorithm for the rotating wedge prism, this technology achieves stable texture appearance and information output, thus alleviating the perception attenuation and insufficient event information problems of traditional event cameras in stationary, slow-moving, or specific motion directions. However, existing AMI-EV artificial microsaccade-enhanced event cameras are monocular configurations, only acquiring two-dimensional visual information. This cannot meet the needs of mobile robots for three-dimensional environmental perception capabilities in tasks such as navigation and obstacle avoidance, including key functions like depth estimation, spatial localization, and obstacle ranging. Meanwhile, the wedge prisms in existing artificial micro-eye movement enhanced event camera systems typically rotate at a preset fixed frequency and cannot be adaptively adjusted according to the camera's motion state. This makes it difficult to balance event triggering efficiency in low-dynamic scenes, compensation error in high-dynamic scenes, and system energy consumption. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an adaptive artificial micro-eye movement event camera binocular system and its control method, which addresses the above-mentioned problems of the prior art. The present invention aims to meet the needs of mobile robots for three-dimensional spatial perception in tasks such as navigation and obstacle avoidance, and can adaptively adjust according to the camera motion state to take into account the event triggering efficiency in low dynamic scenes, the compensation error in high dynamic scenes, and the system energy consumption.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: An adaptive artificial micro-eye-tracking event camera binocular system includes a left event camera, a right event camera, and a mounting base. The left and right event cameras have built-in inertial measurement units (IMUs) for acquiring angular velocity data of motion. The left and right event cameras are mounted parallel to each other on the mounting base and spaced apart at a preset baseline distance. A left lens is located on the front side of the left event camera, and a right lens is located on the front side of the right event camera. A motor is mounted on the mounting base, and its output is connected to a transmission pinion. A rotatable left prism fixing gear and a right prism fixing gear are located on the front side of the left and right event cameras on the mounting base. The left and right prism fixing gears mesh with the transmission pinion to achieve synchronous rotation. A left wedge prism is mounted on the left prism fixing gear, and a right wedge prism is mounted on the right prism fixing gear. A photoelectric switch assembly for detecting a preset reference position of either the left or right prism fixing gear is mounted on the mounting base.

[0005] Optionally, the left prism fixing gear is rotatably mounted on the mounting base plate via a left rotary support assembly, and the right prism fixing gear is rotatably mounted on the mounting base plate via a right rotary support assembly.

[0006] Optionally, the left wedge prism is mounted and fixed on the left prism fixing gear via a left wedge prism mounting base, and the right wedge prism is mounted and fixed on the right prism fixing gear via a right wedge prism mounting base.

[0007] Optionally, the transmission pinion forms a 4:1 transmission ratio with the left prism fixed gear and the right prism fixed gear, respectively.

[0008] Optionally, both the left and right event cameras are DVXplorer event cameras.

[0009] Optionally, the DVXplorer event camera supports multi-camera synchronization and achieves time synchronization through daisy-chain connection and external event injection.

[0010] A control method for the adaptive artificial micro-eye-tracking event camera binocular system includes the following steps: S1, initialize system parameters, the system parameters include a first angular velocity threshold, a second angular velocity threshold, a wedge prism zero position and an initial value of motor feedback, wherein the first angular velocity threshold is used to determine whether the camera is in a low dynamic state, the second angular velocity threshold is used to determine whether the camera is in a high dynamic state, and the first angular velocity threshold is less than the second angular velocity threshold. S2, acquire the angular velocity data detected by the left and right event cameras, as well as the motor feedback signal of the motor; S3. Determine if the angular velocity is less than the first angular velocity threshold. If the angular velocity is less than the first angular velocity threshold, the adaptive artificial micro-eye movement event camera binocular system is determined to be in a low dynamic state. Increase the target speed of the motor to increase the rotation frequency of the left and right wedge prisms, thereby enhancing the artificial micro-eye movement effect and improving the event triggering efficiency. Determine if the angular velocity is greater than the second angular velocity threshold. If the angular velocity is greater than the second angular velocity threshold, the adaptive artificial micro-eye movement event camera binocular system is determined to be in a high dynamic state. Decrease the target speed of the motor to decrease the rotation frequency of the left and right wedge prisms, thereby reducing the additional motion, compensation error, and energy consumption introduced by the wedge prism rotation. Otherwise, maintain the current speed of the motor. S4, based on the determined target speed of the motor, outputs a motor control signal to adjust the rotation state of the left and right wedge prisms, changes the direction of the incident light entering the left and right event cameras, and makes the incident light form an approximately circular motion trajectory on the image plane of the corresponding left and right event cameras, thereby actively triggering event responses at the edge of the scene.

[0011] Optionally, step S4 further includes: obtaining the parallax of the same spatial point in the left and right event views acquired by the left and right event cameras. Based on the parallax in the left and right event views Calculate the depth of the target point: ; in, The depth of the target point. The camera focal lengths of the left and right event cameras. The preset baseline distance between the left and right event cameras.

[0012] Optionally, when initializing system parameters in step S1, the process includes initializing the wedge prism zero position: adjusting the left and right wedge prisms to rotate one revolution by the motor, and detecting the trigger signals generated when the left and right wedge prisms pass through the preset reference position by the photoelectric switch assembly to obtain the wedge prism zero position signal; when acquiring the angular velocity data detected by the left and right event cameras and the motor feedback signal of the motor in step S2, acquiring the motor feedback signal of the motor includes acquiring the motor encoder signal or the motor driver feedback signal, and determining the real-time angles of the left and right wedge prisms based on the wedge prism zero position signal and the motor encoder signal or the motor driver feedback signal.

[0013] Optionally, step S4 further includes event position compensation: based on the determined real-time angles of the left and right wedge prisms, and combined with the pre-calibrated correspondence between the real-time angles of the left and right wedge prisms and the event position offset, the corresponding event position offset is obtained, and the event positions acquired by the left and right event cameras are compensated according to the event position offset, so as to correct the additional motion effects introduced by the rotation of the left and right wedge prisms.

[0014] Compared with the prior art, the present invention can mainly achieve the following beneficial effects: The adaptive artificial micro-eye-tracking event camera binocular system of the present invention drives two wedge prisms to rotate simultaneously through a gear transmission mechanism. The left and right event cameras have built-in inertial measurement units (IMUs) to acquire angular velocity data of the motion. The motion angular velocity can be obtained from the built-in IMU of the event camera. The adaptive micro-eye-tracking control module dynamically adjusts the target speed of the drive motor according to the angular velocity and controls the drive motor to drive the wedge prisms to rotate. This can meet the needs of mobile robots for three-dimensional spatial perception in navigation, obstacle avoidance and other tasks. It can also adaptively adjust according to the camera motion state to take into account the event triggering efficiency in low dynamic scenes, the compensation error in high dynamic scenes and the system energy consumption. Attached Figure Description

[0015] Figure 1 This is a stereo exploded view of the adaptive artificial micro-eye movement event camera binocular system in an embodiment of the present invention.

[0016] Figure 2 This is a three-dimensional structural diagram of the adaptive artificial micro-eye movement event camera binocular system in an embodiment of the present invention.

[0017] Figure 3 This is a schematic diagram of the control flow of the adaptive artificial micro-eye movement event camera binocular system in an embodiment of the present invention.

[0018] Legend: 1. Left event camera; 2. Right event camera; 3. Left lens; 4. Right lens; 5. Motor; 6. Mounting base plate; 7. Left prism fixing gear; 8. Right prism fixing gear; 9. Left wedge prism; 10. Right wedge prism; 11. Transmission pinion; 12. Photoelectric switch assembly; 13. Left rotary support assembly; 14. Right rotary support assembly; 15. Left wedge prism mounting base; 16. Right wedge prism mounting base. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be further described in detail below with reference to the accompanying drawings in the embodiments of the present invention.

[0020] like Figure 1 and Figure 2As shown, the adaptive artificial micro-eye movement event camera binocular system of this embodiment includes a left event camera 1, a right event camera 2, and a mounting plate 6. The left event camera 1 and right event camera 2 have built-in inertial measurement units (IMUs) for acquiring angular velocity data of the motion. The left event camera 1 and right event camera 2 are mounted parallel to each other on the mounting plate 6 and spaced apart at a preset baseline distance, for example, 100mm. The left event camera 1 has a left lens 3 on its front side, and the right event camera 2 has a right lens 4 on its front side. A motor 5 is mounted on the mounting plate 6, and the output end of the motor 5 is connected to a transmission pinion 1. 1. A rotatable left prism fixing gear 7 and right prism fixing gear 8 are provided on the mounting base plate 6 in front of the left event camera 1 and the right event camera 2. The left prism fixing gear 7 and right prism fixing gear 8 respectively mesh with the transmission pinion 11 to achieve synchronous rotation. A left wedge prism 9 is installed on the left prism fixing gear 7, and a right wedge prism 10 is installed on the right prism fixing gear 8. A photoelectric switch assembly 12 is installed on the mounting base plate 6 to detect the preset reference position of the left prism fixing gear 7 or the right prism fixing gear 8, and to detect the preset reference position of the rotating component. In this embodiment, the adaptive artificial micro-eye-tracking event camera binocular system achieves three-dimensional environmental perception by setting rotatable wedge prisms in front of the apertures of the two event cameras, combined with the principle of binocular parallax, effectively overcoming the limitation of monocular systems that can only acquire two-dimensional information. This embodiment of the adaptive artificial micro-eye movement event camera binocular system utilizes the principle of binocular parallax to achieve depth estimation, upgrading the system from two-dimensional perception to three-dimensional perception and expanding the application potential of event cameras in scenarios requiring spatial perception, such as mobile robot navigation. The micro-eye movement rotation frequency is dynamically adjusted according to the camera's motion state. During rapid movement, the rotation speed is reduced to minimize compensation errors, while during smooth movement or stillness, the rotation speed is increased to maintain event triggering efficiency, achieving a balance between performance and efficiency.

[0021] like Figure 1 As shown, in this embodiment, the left prism fixing gear 7 is rotatably mounted on the mounting base plate 6 via the left rotary support assembly 13, and the right prism fixing gear 8 is rotatably mounted on the mounting base plate 6 via the right rotary support assembly 14.

[0022] like Figure 1 As shown, in this embodiment, the left wedge prism 9 is mounted and fixed on the left prism fixing gear 7 via the left wedge prism mounting base 15, and the right wedge prism 10 is mounted and fixed on the right prism fixing gear 8 via the right wedge prism mounting base 16.

[0023] The transmission ratio between the transmission pinion 11 and the left prism fixed gear 7 can be configured as needed. For example, in this embodiment, the transmission ratio between the transmission pinion 11 and the left prism fixed gear 7 and the right prism fixed gear 8 is 4:1. Specifically, in this embodiment, the transmission pinion 11 has 20 teeth, and the left prism fixed gear 7 and the right prism fixed gear 8 have 80 teeth.

[0024] In this embodiment, both the left event camera 1 and the right event camera 2 are DVXplorer event cameras with a pixel resolution of 640×480 and dimensions of 40×60×25mm. Each event camera has a built-in inertial measurement unit (IMU) for acquiring the camera's angular velocity. In this embodiment, the DVXplorer event cameras support multi-camera synchronization and achieve time synchronization through daisy-chain connections and external event injection.

[0025] like Figure 3 As shown, this embodiment also provides a control method for the aforementioned adaptive artificial micro-eye-tracking event camera binocular system, including the following steps: S1, initialize system parameters, the system parameters include a first angular velocity threshold, a second angular velocity threshold, a wedge prism zero position and an initial value of motor feedback, wherein the first angular velocity threshold is used to determine whether the camera is in a low dynamic state, the second angular velocity threshold is used to determine whether the camera is in a high dynamic state, and the first angular velocity threshold is less than the second angular velocity threshold. S2, acquire the angular velocity data detected by the left event camera 1 and the right event camera 2, as well as the motor feedback signal of the motor 5; S3. Determine if the angular velocity is less than the first angular velocity threshold. If the angular velocity is less than the first angular velocity threshold, the adaptive artificial micro-eye movement event camera binocular system is determined to be in a low dynamic state. Increase the target speed of motor 5 to increase the rotation frequency of the left wedge prism 9 and the right wedge prism 10, thereby enhancing the artificial micro-eye movement effect and improving the event triggering efficiency. Determine if the angular velocity is greater than the second angular velocity threshold. If the angular velocity is greater than the second angular velocity threshold, the adaptive artificial micro-eye movement event camera binocular system is determined to be in a high dynamic state. Decrease the target speed of motor 5 to decrease the rotation frequency of the left wedge prism 9 and the right wedge prism 10, thereby reducing the additional motion, compensation error and energy consumption introduced by the wedge prism rotation. Otherwise, maintain the current speed of motor 5. S4. Based on the determined target speed of motor 5, output motor control signal to adjust the rotation state of left wedge prism 9 and right wedge prism 10, change the direction of incident light entering left event camera 1 and right event camera 2, so that the incident light forms an approximately circular motion trajectory on the image plane of the corresponding left event camera 1 and right event camera 2, thereby actively triggering event response at the edge of the scene.

[0026] This embodiment's adaptive artificial micro-eye-tracking event camera binocular system control method dynamically adjusts the prism rotation frequency based on the real-time acquisition of the motion angular velocity of the event camera's built-in IMU. When the angular velocity is high, the frequency is reduced to minimize compensation errors; when the angular velocity is low, the frequency is increased to ensure event triggering efficiency, achieving an adaptive balance between performance and efficiency. A first angular velocity threshold is used to determine whether the event camera or robot platform is in a low-dynamic state: when the angular velocity output by the IMU is less than the first threshold, it indicates that the camera's own motion is weak, and external motion contributes little to event triggering. Therefore, the target rotation speed of the wedge prism is increased to enhance the artificial micro-eye-tracking effect. A second angular velocity threshold is used to determine whether the event camera or robot platform is in a high-dynamic state: when the angular velocity output by the IMU is greater than the second threshold, it indicates that the camera's own motion is strong, and the event camera can generate more events through its own motion. Therefore, the target rotation speed of the wedge prism is reduced to decrease the compensation errors and energy consumption introduced by the wedge prism rotation. Therefore, in this embodiment, after acquiring the IMU angular velocity data and motor feedback signals, the current angular velocity is compared with the first and second angular velocity thresholds, respectively. When the angular velocity is less than the first angular velocity threshold, the target rotational speed is increased; when the angular velocity is greater than the second angular velocity threshold, the target rotational speed is decreased; when the angular velocity is between the first and second angular velocity thresholds, the current rotational speed is maintained. Then, a motor control signal is output to drive the left and right wedge prisms to rotate, and the above steps are repeated cyclically.

[0027] Two event cameras, left event camera 1 and right event camera 2, respectively acquire left and right event stream data. Depth estimation is performed using the binocular parallax principle to obtain three-dimensional spatial information of the environment. For example, step S4 in this embodiment further includes: acquiring the parallax of the same spatial point in the left and right event views acquired by left event camera 1 and right event camera 2. Based on the parallax in the left and right event views Calculate the depth of the target point: ; in, The depth of the target point. The focal lengths of the left event camera 1 and the right event camera 2 are given. The preset baseline distance between the left event camera 1 and the right event camera 2 (100 mm in this embodiment).

[0028] In step S1 of this embodiment, when initializing system parameters, the initialization of the wedge prism zero position includes: adjusting the left wedge prism 9 and the right wedge prism 10 to rotate one revolution by the motor 5, and detecting the trigger signal generated when the left wedge prism 9 and the right wedge prism 10 pass through the preset reference position by the photoelectric switch assembly 12 to obtain the wedge prism zero position signal; in step S2, when obtaining the angular velocity data detected by the left event camera 1 and the right event camera 2, and the motor feedback signal of the motor 5, obtaining the motor feedback signal of the motor 5 includes obtaining the motor encoder signal or the motor driver feedback signal, and determining the real-time angle of the left wedge prism 9 and the right wedge prism 10 according to the wedge prism zero position signal and the motor encoder signal or the motor driver feedback signal.

[0029] In this embodiment, when the left wedge prism 9 and the right wedge prism 10 rotate, they change the direction of the incident light rays entering the left event camera 1 and the right event camera 2, respectively, so that the incident light rays form an approximately circular motion trajectory on the image plane of the corresponding event camera, thereby actively triggering the event response at the edge of the scene. The system can establish the correspondence between the rotation angle of the wedge prism and the offset of the event position through wedge prism orientation calibration, and obtain the real-time angle of the wedge prism according to the zero-position signal output by the photoelectric switch component 12 to compensate for the event position and correct the additional motion effect introduced by the rotation of the wedge prism. Step S4 in this embodiment also includes event position compensation: based on the determined real-time angles of the left wedge prism 9 and the right wedge prism 10, and combined with the pre-calibrated correspondence between the real-time angles of the left wedge prism 9 and the right wedge prism 10 and the event position offset, the corresponding event position offset is obtained. Based on the event position offset, the event positions collected by the left event camera 1 and the right event camera 2 are compensated respectively to correct the additional motion effects introduced by the rotation of the left wedge prism 9 and the right wedge prism 10.

[0030] The control method of the adaptive artificial micro-eye-tracking event camera binocular system in this embodiment can obtain three-dimensional spatial information of the environment from the left and right event stream data, which can be used for obstacle distance estimation, spatial perception, navigation or obstacle avoidance of mobile robots.

[0031] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. An adaptive artificial micro-eye movement event camera binocular system, characterized in that, The system includes a left event camera (1), a right event camera (2), and a mounting plate (6). The left event camera (1) and right event camera (2) are equipped with built-in inertial measurement units (IMUs) to acquire angular velocity data of the motion. The left event camera (1) and right event camera (2) are mounted parallel to each other on the mounting plate (6) and spaced at preset baseline intervals. The left event camera (1) has a left lens (3) on its front side, and the right event camera (2) has a right lens (4) on its front side. A motor (5) is mounted on the mounting plate (6), and the output end of the motor (5) is connected to a transmission pinion (11). On the plate (6), in front of the left event camera (1) and the right event camera (2), there are rotatable left prism fixed gear (7) and right prism fixed gear (8). The left prism fixed gear (7) and right prism fixed gear (8) mesh with the transmission pinion (11) to achieve synchronous rotation. A left wedge prism (9) is installed on the left prism fixed gear (7) and a right wedge prism (10) is installed on the right prism fixed gear (8). A photoelectric switch assembly (12) for detecting the preset reference position of the left prism fixed gear (7) or the right prism fixed gear (8) is installed on the mounting base plate (6).

2. The adaptive artificial micro-eye movement event camera binocular system according to claim 1, characterized in that, The left prism fixed gear (7) is rotatably mounted on the mounting base plate (6) via the left rotary support assembly (13), and the right prism fixed gear (8) is rotatably mounted on the mounting base plate (6) via the right rotary support assembly (14).

3. The adaptive artificial micro-eye movement event camera binocular system according to claim 2, characterized in that, The left wedge prism (9) is mounted and fixed on the left prism fixing gear (7) via the left wedge prism mounting base (15), and the right wedge prism (10) is mounted and fixed on the right prism fixing gear (8) via the right wedge prism mounting base (16).

4. The adaptive artificial micro-eye movement event camera binocular system according to claim 1, characterized in that, The transmission pinion (11) forms a transmission ratio of 4:1 with the left prism fixed gear (7) and the right prism fixed gear (8).

5. The adaptive artificial micro-eye movement event camera binocular system according to claim 1, characterized in that, Both the left event camera (1) and the right event camera (2) are DVXplorer event cameras.

6. The adaptive artificial micro-eye movement event camera binocular system according to claim 5, characterized in that, The DVXplorer event camera supports multi-camera synchronization and achieves time synchronization through daisy chain connection and external event injection.

7. A control method for an adaptive artificial micro-eye-tracking event camera binocular system according to any one of claims 1 to 6, characterized in that, Includes the following steps: S1, initialize system parameters, the system parameters include a first angular velocity threshold, a second angular velocity threshold, a wedge prism zero position and an initial value of motor feedback, wherein the first angular velocity threshold is used to determine whether the camera is in a low dynamic state, the second angular velocity threshold is used to determine whether the camera is in a high dynamic state, and the first angular velocity threshold is less than the second angular velocity threshold. S2, acquire the angular velocity data detected by the left event camera (1) and the right event camera (2), as well as the motor feedback signal of the motor (5); S3, determine whether the angular velocity is less than the first angular velocity threshold. If the angular velocity is less than the first angular velocity threshold, it is determined that the adaptive artificial micro-eye movement event camera binocular system is in a low dynamic state. Increase the target speed of the motor (5) to increase the rotation frequency of the left wedge prism (9) and the right wedge prism (10), thereby enhancing the artificial micro-eye movement effect and improving the event triggering efficiency. Determine whether the angular velocity is greater than the second angular velocity threshold. If the angular velocity is greater than the second angular velocity threshold, it is determined that the adaptive artificial micro-eye movement event camera binocular system is in a high dynamic state. Decrease the target speed of the motor (5) to decrease the rotation frequency of the left wedge prism (9) and the right wedge prism (10), thereby reducing the additional motion, compensation error and energy consumption introduced by the rotation of the wedge prism. Otherwise, maintain the current speed of the motor (5). S4, according to the determined target speed of the motor (5), output motor control signal to adjust the rotation state of the left wedge prism (9) and the right wedge prism (10), change the direction of the incident light entering the left event camera (1) and the right event camera (2), so that the incident light forms an approximately circular motion trajectory on the image plane of the corresponding left event camera (1) and right event camera (2), thereby actively triggering the event response at the edge of the scene.

8. The control method for the adaptive artificial micro-eye-tracking event camera binocular system according to claim 7, characterized in that, Step S4 also includes: acquiring the parallax of the same spatial point in the left and right event views acquired by the left event camera (1) and the right event camera (2). Based on the parallax in the left and right event views Calculate the depth of the target point: ； in, The depth of the target point. The camera focal lengths of the left event camera (1) and the right event camera (2) are given. The preset baseline distance between the left event camera (1) and the right event camera (2).

9. The control method for the adaptive artificial micro-eye-tracking event camera binocular system according to claim 7, characterized in that, When initializing system parameters in step S1, the wedge prism zero position is initialized: the left wedge prism (9) and the right wedge prism (10) are rotated one revolution by the motor (5), and the trigger signal generated when the left wedge prism (9) and the right wedge prism (10) pass through the preset reference position is detected by the photoelectric switch assembly (12) to obtain the wedge prism zero position signal; when obtaining the angular velocity data detected by the left event camera (1) and the right event camera (2) and the motor feedback signal of the motor (5) in step S2, the motor feedback signal of the motor (5) is obtained by obtaining the motor encoder signal or the motor driver feedback signal, and the real-time angle of the left wedge prism (9) and the right wedge prism (10) is determined according to the wedge prism zero position signal and the motor encoder signal or the motor driver feedback signal.

10. The control method for the adaptive artificial micro-eye-tracking event camera binocular system according to claim 7, characterized in that, Step S4 also includes event position compensation: based on the determined real-time angles of the left wedge prism (9) and the right wedge prism (10), combined with the pre-calibrated correspondence between the real-time angles of the left wedge prism (9) and the right wedge prism (10) and the event position offset, the corresponding event position offset is obtained, and the event positions collected by the left event camera (1) and the right event camera (2) are compensated according to the event position offset, so as to correct the additional motion effects introduced by the rotation of the left wedge prism (9) and the right wedge prism (10).

Images