Simulated driving vision training system for low-vision people

Abstract

The invention provides a simulated driving visual training system for low-vision people. The simulated driving visual training system is used for training low-vision drivers to detect roadside static pedestrians, approaching pedestrians and intersection static pedestrians on the blind side and the visible side in a simulated driving scene. The system can be applied to the fields of computer science, psychology and medicine, simulates the driving condition in a virtual driving environment, and trains and evaluates the detection capability of a low-vision driver on roadside pedestrians. And various data in the driving process are recorded for subsequent evaluation and analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality and visual simulation, and in particular to a simulated driving vision training system for people with low vision. Background Art

[0002] Traditional visual impairment simulation driving technology often has the following defects:

[0003] Limited scene authenticity: Traditional visually impaired simulation driving technologies often lack highly realistic three-dimensional virtual scenes and customizable driving kit interactive control functions, which limits the authenticity and developability of the user experience.

[0004] Limitations of pedestrian simulation: Existing systems are limited in the types and behaviors of pedestrians they can simulate, and are unable to simulate and simulate multiple pedestrian scenarios. In addition, these systems lack the ability to be customized based on actual traffic conditions, and are unable to provide a more comprehensive detection training experience.

[0005] Lack of simulation of eye-attracting events: Existing visually impaired driving simulation systems generally lack the ability to simulate eye-attracting events. This limitation leads to poor training and testing results in complex driving environments. Existing systems fail to effectively integrate simulation of eye-attracting events such as oncoming vehicles and following vehicles, and are unable to simulate the complexity of pedestrian detection tasks in real scenarios, making it difficult to improve users' detection and reaction capabilities in complex driving scenarios.

[0006] Limitations of response recording: The existing system does not fully integrate the driving kit for real-time response recording, resulting in insufficient accuracy in the collection of user operation data (such as steering wheel rotation, button action, and pedal use). It is impossible to accurately match these response data with the time and location of pedestrians, which affects the evaluation of user detection ability, reaction speed, and driving stability.

[0007] Limitations in comprehensive collection of driving data: The existing system fails to provide a complete driving data collection function, and it is difficult to fully record dynamic data during driving, including timestamps, car positions, car speeds, pedestrian initial positions, pedestrian running trajectories, pedestrian running speeds, steering wheel rotation data, steering wheel button data, pedal data, and other information, which limits the subsequent comprehensive analysis and research of driving behavior.

[0008] Lack of a reasonable simulated driving vision training system: The existing system cannot accurately simulate complex visual scenes and lacks the ability to simulate various visual interferences and complex lighting conditions. This limits the driver's adaptive training in a changing environment, making it impossible for the driver to be fully prepared to face the various visual challenges in actual driving, affecting the improvement of comprehensive driving skills. Summary of the invention

[0009] According to the technical problems mentioned in the above background technology, a simulated driving vision training system for people with low vision is provided.

[0010] The technical means adopted by the present invention are as follows:

[0011] A simulated driving vision training system for people with low vision, comprising:

[0012] Driving interaction control module, dynamic pedestrian simulation module, eye-attracting event simulation module, response recording module, simulated driving vision training module and data storage and analysis module;

[0013] The driving interactive control module uses 3D virtual reality technology to develop an operating interface that is compatible with the Logitech driving kit, enabling users to accurately control the virtual vehicle and customize personalized functions;

[0014] The dynamic pedestrian simulation module dynamically simulates the behaviors and interaction modes of various pedestrians in the virtual scene based on preset positions, distances, angles and speeds through 3D modeling and animation technology;

[0015] The eye-attracting event simulation module increases the complexity of the pedestrian detection task by simulating a variety of eye-attracting events to achieve a realistic driving scenario;

[0016] The response recording module obtains the user's response data during driving through the Logitech Driving Kit, which corresponds to the time and location of the pedestrian's appearance;

[0017] The simulated driving vision training module is used to train low-vision drivers’ detection ability and reaction speed to stationary and approaching pedestrians at different eccentric distances, as well as stationary pedestrians at intersections, and the data collected during the training is used for subsequent analysis;

[0018] The data storage and analysis module is used to store and analyze the car behavior data, pedestrian dynamic data and driving interaction data generated during the simulated driving process, and comprehensively evaluate the driving ability of people with low vision, their visual detection ability of pedestrians, reaction time and driving stability;

[0019] The driving interaction control module drives the user to control the virtual vehicle to respond to the complex pedestrian detection task by linking with the dynamic pedestrian simulation module and the sight-attracting event simulation module;

[0020] The response recording module acquires the interaction data of the driving interaction control module in real time, and combines the data of the dynamic pedestrian simulation module and the eye-attracting event simulation module for subsequent evaluation of the low-vision user's ability to detect pedestrians, reaction speed, and driving stability;

[0021] The data storage and analysis module includes vehicle behavior data, pedestrian dynamic data, and user response data, and analyzes key indicators.

[0022] Furthermore, the vehicle behavior data includes real-time speed, position and collision information.

[0023] Furthermore, the pedestrian dynamic data includes initial position, movement speed, movement trajectory and interaction events.

[0024] Furthermore, the data storage and analysis module stores all data during the driving process at a frequency of 30 Hz.

[0025] Furthermore, the events that attract attention include oncoming vehicles, following vehicles, vehicles coming from intersections, and vehicles coming from behind.

[0026] Furthermore, the response data includes steering wheel rotation, button operation and pedal data.

[0027] Compared with the prior art, the present invention has the following advantages:

[0028] The present invention provides a simulated driving vision training system for people with low vision, which is used to train low vision drivers to detect roadside stationary pedestrians, approaching pedestrians and stationary pedestrians at intersections on the blind side and visible side in a simulated driving scene.

[0029] The driving interactive control module has developed an operating interface for using the driving kit to accurately control the virtual vehicle, allowing users to complete steering, acceleration, braking and other operations in the virtual scene. Low-vision drivers can experience driving on urban and rural roads by controlling the vehicle. The module also includes simulation of the vehicle's physical characteristics to ensure the vehicle's true response under different road conditions and improve the effectiveness of driving training. In addition, it also supports user personalized function customization, which improves the interactivity and applicability of user training.

[0030] The dynamic pedestrian simulation module uses 3D modeling and animation technology to simulate pedestrian behavior and interaction in different situations. Pedestrians in various motion states (stationary, walking, running) appear at small azimuths (4°) and large azimuths (14°) on both sides of the road and at intersections. The module supports customizing pedestrian appearance and behavior patterns, such as pedestrian height, clothing color, movement mode, etc. Combined with collision detection technology, it responds to the interaction between pedestrians and vehicles in real time, improving the flexibility and authenticity of training.

[0031] The attention-attracting event simulation module simulates attention-attracting events including oncoming vehicles, following vehicles, vehicles at intersections, and vehicles coming from behind. Oncoming vehicles simulate oncoming traffic in actual driving, the rear vehicle scenario simulates the approach of the rear vehicle, and the intersection vehicle scenario simulates the side vehicle entering the main road. These scenarios increase the risk of driver distraction and enhance the realism and challenge of driving training.

[0032] The response recording module uses the Logitech G923 driving kit to obtain the user's response data during driving and displays it on the screen in real time, including steering wheel rotation, button operation, and pedal action, which accurately corresponds to the time and location of pedestrians. The response data is displayed in real time on the driving interface and is subsequently used to evaluate the driver's detection ability and reaction speed, providing reliable data support for driving behavior research.

[0033] The simulated driving vision training module is used to train low-vision drivers in urban and rural scenes, at different driving speeds (48 km / h and 96 km / h) to detect stationary and approaching pedestrians at different eccentric distances (4° and 14°), as well as stationary pedestrians at intersections. Through multiple training sessions, low-vision drivers can improve their detection ability, reaction speed and driving stability for pedestrians on the blind side and visible side.

[0034] The data storage and analysis module systematically stores and manages various data generated during the simulated driving process, including the real-time speed, position, collision data of the vehicle, the appearance time, initial position, movement trajectory, movement speed and disappearance time of pedestrians, as well as the data of the eye-catching event and response recording module. Combined with the timestamp, the driver's detection ability of stationary pedestrians, approaching pedestrians and stationary pedestrians at intersections at different eccentric distances is evaluated, and key indicators such as the non-detection ratio, reaction time, collision risk and driving stability are analyzed. It provides comprehensive and reliable data support for driving behavior research and training program improvement, while ensuring efficient data retrieval and multi-dimensional analysis.

[0035] In summary, the system of this application can be applied to the fields of computer science, psychology, and medicine, simulating driving conditions in a virtual driving environment, and training and evaluating the ability of low-vision drivers to detect roadside pedestrians. Various data during driving can be recorded for subsequent evaluation and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0037] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0038] Figure 2 It is a data flow diagram between modules of the present invention.

[0039] Figure 3 It is an overall schematic diagram of the positions of stationary pedestrians on the roadside, running pedestrians and pedestrians at a three-way intersection according to the present invention.

[0040] Figure 4 It is a schematic diagram of an oncoming vehicle according to the present invention.

[0041] Figure 5 It is a schematic diagram of following a vehicle according to the present invention.

[0042] Figure 6 It is a schematic diagram of a vehicle coming from behind according to the present invention.

[0043] Figure 7 This is a schematic diagram of vehicles approaching at an intersection according to the present invention. DETAILED DESCRIPTION

[0044] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0045] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, system, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, systems, products or devices.

[0046] like Figure 1-Figure 7 As shown, the present invention provides a simulated driving vision training system for people with low vision, comprising:

[0047] Driving interaction control module, dynamic pedestrian simulation module, eye-attracting event simulation module, response recording module, simulated driving vision training module and data storage and analysis module;

[0048] The driving interactive control module enables users to achieve precise control of the vehicle in a virtual scene, including steering, acceleration, braking and other operations, by developing an operating interface that is compatible with the driving kit. The maximum speed of the vehicle is limited to 48 km / h and 96 km / h in urban environments and rural highways, respectively, which is determined based on actual driving conditions, the location of pedestrians and the maximum reaction time of 5 seconds. Through the comprehensive application of advanced virtual reality technology and precise physical simulation, this module provides users with a better dynamic driving experience and can effectively improve the training effect of low-vision drivers.

[0049] In this application, the dynamic pedestrian simulation module simulates pedestrian behavior in different situations based on preset positions, distances, angles and speeds through 3D modeling technology and animation technology. Pedestrians are set to be about 2 meters tall, wearing white shirts and blue pants, and appear on the right or left side of the road every 15-60 seconds.

[0050] In urban and rural road scenes, pedestrians can appear in many ways, such as stationary pedestrians on the roadside, walking pedestrians, running pedestrians, and stationary pedestrians at three-way intersections. The whole process from appearance to disappearance is simulated in detail. The stationary pedestrians on the roadside suddenly appear and remain stationary, with the initial position located 67 meters in front of the vehicle, the lateral distance is 4.7 meters, the horizontal angle is 4°, and the vertical angle is 1.78°; the running pedestrians approach the road from both sides at a speed of 3.34m / s, with the initial position located 67 meters in front of the vehicle, the lateral distance is 16.7 meters, the horizontal angle is 14°, and the vertical angle is 1.78°. Walking and running pedestrians always disappear before colliding with the car to avoid collision; stationary pedestrians at the three-way intersection appear near the zebra crossing on the left, right, and opposite sides.

[0051] The dynamic pedestrian simulation module also supports the customization of pedestrian behavior and appearance according to specific scene requirements, including pedestrian height, clothing color, and interactive functions, to improve the flexibility of simulation. By combining collision detection technology in virtual scenes, this module can respond to the interaction between pedestrians and vehicles in real time. In addition, the dynamic pedestrian simulation module can be integrated with other modules (such as the eye-attracting event simulation module and the response recording module) to improve the comprehensive effect of driving training. For example, pedestrians may suddenly appear when an eye-attracting event occurs, further increasing the detection difficulty and training intensity of low-vision patients. All relevant data of pedestrians will be recorded in detail for subsequent analysis of the reaction speed and detection ability of low-vision drivers, providing a comprehensive performance evaluation. Through highly realistic and dynamic pedestrian simulation, it helps low-vision patients train and improve their detection ability and reaction speed for pedestrians on the blind side and visible side in actual driving situations.

[0052] Preferably, in the present application, the sight-attracting event simulation module simulates a variety of sight-attracting events, including oncoming vehicles, following vehicles, vehicles coming from intersections, and vehicles coming from behind. The simulated vehicle types include ordinary blue vehicles and police cars with flashing lights.

[0053] Oncoming vehicles will come towards you at a set speed during driving, simulating the oncoming traffic in actual driving. Oncoming vehicles not only increase the risk of distraction for low-vision drivers, but also require drivers to adjust lanes in time to avoid collisions, which increases the realism and complexity of driving training. Figure 4 shown.

[0054] In the following vehicle driving scenario, the driver needs to maintain an appropriate distance and pay attention to the deceleration or lane change of the vehicle in front. The following vehicle can be an ordinary blue vehicle or a police car with flashing lights. Through this scenario simulation, the low-vision driver's ability to maintain a safe distance under complex traffic conditions is trained. Figure 5 shown.

[0055] The rear vehicle scenario simulates a situation where a rear vehicle approaches at a set speed. The driver needs to observe the movement of the rear vehicle through the rearview mirror to ensure safe driving. This scenario effectively trains low-vision drivers' all-round observation ability in complex traffic environments and helps low-vision drivers better deal with the approach of rear vehicles in real driving. Figure 6 shown.

[0056] By adjusting the appearance time and location of the above vehicles, each driving training scenario is ensured to be unique and challenging. For example, vehicles may suddenly appear at different times and locations, forcing the driver to react in a short time. All event data will be recorded in detail for subsequent driver reaction analysis and evaluation.

[0057] Oncoming vehicles at an intersection simulate vehicles that suddenly appear at an intersection. These vehicles may come from the side roads and enter the main road at a set speed, requiring the driver to make judgments and react in a short time to avoid collision. The addition of this scenario further increases the complexity and challenge of the driving environment. Figure 7 shown.

[0058] By simulating a variety of eye-catching events, this module effectively increases the complexity and realism of driving tasks, helping low-vision drivers improve their ability to detect pedestrians in actual driving. The comprehensive application of this module ensures that users can experience highly realistic driving challenges in different driving scenarios, thereby significantly improving training results.

[0059] Preferably, the response recording module acquires steering wheel rotation data, steering wheel button data, pedal data, baffle data, etc. in real time, and accurately matches these data with the time of pedestrian appearance and eye-attracting event occurrence according to the timestamp, for subsequent evaluation of the low-vision driver's detection ability and reaction speed to pedestrians on the blind side and visible side.

[0060] During driving, the user operates the steering wheel and pedals according to the voice prompts of the scene, such as turning left, turning right, going straight, entering a roundabout, and exiting a roundabout. From the beginning to the end of each driving task, the steering wheel rotation data and pedal pressing data are obtained in real time.

[0061] During driving, pressing a specific button on the steering wheel indicates that the user has spotted a pedestrian. The exact time when the user presses the button is obtained and compared with the time when the pedestrian appears, which is used to calculate the reaction time later.

[0062] During system operation, all response data corresponding to this module are displayed in real time in the upper left corner of the driving interface.

[0063] The simulated driving vision training module is used to train low-vision drivers to detect stationary and approaching pedestrians that randomly appear at the minimum angle of eccentricity (4°) and the maximum angle of eccentricity (14°), as well as stationary pedestrians at intersections;

[0064] The data storage and analysis module first stores various types of data generated during the simulated driving process; for the driving interaction control module, the real-time speed, position and collision data of the vehicle are recorded.

[0065] For the dynamic pedestrian simulation module, whether it is a stationary pedestrian on the roadside, a walking pedestrian, a running pedestrian or a stationary pedestrian at a three-way intersection, their initial position, appearance time, movement speed, movement trajectory and disappearance time are all accurately recorded.

[0066] For the attention-attracting event simulation module, the type of vehicle (such as a blue ordinary vehicle or a police car with flashing lights), initial position, appearance time, movement speed, movement trajectory and interaction with the low-vision driver are recorded.

[0067] For the response recording module, the steering wheel rotation data, steering wheel button data, pedal data, baffle data and timestamp are recorded.

[0068] The data storage and analysis module combines the above data with timestamps to evaluate the driver's detection ability for stationary pedestrians, approaching pedestrians and stationary pedestrians at intersections at different eccentric distances, and analyzes key indicators such as the proportion of undetected pedestrians, reaction time, collision risk and driving stability.

[0069]

[0070] Among them, the number of delayed reaction pedestrians refers to the number of pedestrians that the driver actually detected but whose reaction time exceeded the maximum safe reaction time.

[0071] Driving stability is expressed by average path deviation, which is divided into straight sections and turning sections. The average path deviation calculation system for straight sections is:

[0072]

[0073] Among them, ALD represents the average path deviation, N represents the total number of sampling points in a certain section, and x i represents the lateral position coordinate of the vehicle at the i-th sampling point, x c is the lateral position coordinate of the lane centerline. In straight sections, only the lateral position difference needs to be considered.

[0074] For curved road sections, the vertical distance between the vehicle and the centerline of the lane needs to be considered when calculating the average path offset, as shown in the following formula:

[0075]

[0076] Among them, (x i, y i ) represents the actual position coordinates of the vehicle at the i-th sampling point, (x c, y c ) is the position coordinate of the point closest to the lane centerline of the i-th sampling point.

[0077] By quantifying detection capabilities, collision risks and driving stability, it provides comprehensive and reliable data support for driving behavior research and training program improvements, helping people with low vision improve the effectiveness of simulated driving vision training.

[0078] After the driving interaction control module successfully controls the virtual vehicle through the driving kit, various behaviors and interaction methods of pedestrians are established through the dynamic pedestrian simulation module, and then various attention-attracting events are simulated through the attention-attracting event simulation module. The response recording module obtains relevant data of the driving interaction control module in real time, and combines the data of the dynamic pedestrian simulation module and the attention-attracting event simulation module to reflect the response of the low-vision driver during the driving process, and evaluates the detection ability and reaction speed of the low-vision driver to pedestrians on the blind side and the visible side; a fixed frequency storage module of 30 Hz is used to store and manage various types of data generated during the simulated driving process and analyze them.

[0079] Embodiment 1:

[0080] As an embodiment of the present application, as a simulated driving vision training system: before each test begins, the user needs some time to familiarize himself with and practice the driving simulator (approximately 30-45 minutes), and then completes five test drives of approximately 10 minutes. Participants are required to obey all normal road rules, try to maintain the prescribed driving speed, and press the designated button on the steering wheel when they see pedestrians. Pedestrians appear on the right or left side of the road once every 15-60 seconds, with a small (approximately 4°, stationary pedestrians) or large (approximately 14°, moving pedestrians) eccentricity relative to the direction of travel of the car.

[0081] The maximum speed of the vehicle in the urban environment and rural highway is limited to 48 km / h and 96 km / h respectively, and the initial position of the pedestrian is 67 meters and 134 meters in front of the vehicle. When driving at the maximum speed, it is equivalent to a distance of 5 seconds, which is twice the 2.5-second perception-braking time of the minimum recommended stopping sight distance in safe road design. When the car approaches the running pedestrian, it calculates whether it will collide based on the car's real-time speed, braking acceleration and distance. If it is expected to hit the running pedestrian, the pedestrian is made to disappear to avoid the collision.

[0082] The detection performance of 52 stationary pedestrians and 52 approaching pedestrians appearing on a straight road segment was analyzed. The main dependent variables were the proportion of non-detections (the number of non-detections as a proportion of the total number of pedestrian events) and the reaction time to detection (the delay from the appearance of the pedestrian to the horn being honked). The reaction to each pedestrian appearance was classified as a missed detection (not seen), a delayed reaction, or a timely reaction, taking into account the reaction time, the car speed, and the assumed braking time required to stop the car. A delayed reaction refers to a situation where the reaction time is too long and there is not enough time to stop; a timely reaction is when the driver sees the pedestrian and has enough time to stop the car. Miss detections and delayed reactions are collectively referred to as untimely reactions and represent potential collisions.

[0083] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A simulated driving vision training system for people with low vision, characterized in that: include: Driving interaction control module, dynamic pedestrian simulation module, eye-attracting event simulation module, response recording module, simulated driving vision training module and data storage and analysis module; The driving interactive control module uses 3D virtual reality technology to develop an operating interface that is compatible with the Logitech driving kit, enabling users to accurately control the virtual vehicle and customize personalized functions; The dynamic pedestrian simulation module dynamically simulates the behaviors and interaction modes of various pedestrians in the virtual scene based on preset positions, distances, angles and speeds through 3D modeling and animation technology; The eye-attracting event simulation module increases the complexity of the pedestrian detection task by simulating a variety of eye-attracting events to achieve a realistic driving scenario; The response recording module obtains the user's response data during driving through the Logitech Driving Kit, which corresponds to the time and location of the pedestrian's appearance; The simulated driving vision training module is used to train low-vision drivers’ detection ability and reaction speed to stationary and approaching pedestrians at different eccentric distances, as well as stationary pedestrians at intersections, and the data collected during the training is used for subsequent analysis; The data storage and analysis module is used to store and analyze the car behavior data, pedestrian dynamic data and driving interaction data generated during the simulated driving process, and comprehensively evaluate the driving ability of people with low vision, their visual detection ability of pedestrians, reaction time and driving stability; The driving interaction control module drives the user to control the virtual vehicle to respond to the complex pedestrian detection task by linking with the dynamic pedestrian simulation module and the sight-attracting event simulation module; The response recording module acquires the interaction data of the driving interaction control module in real time, and combines the data of the dynamic pedestrian simulation module and the eye-attracting event simulation module for subsequent evaluation of the low-vision user's ability to detect pedestrians, reaction speed, and driving stability; The data storage and analysis module includes vehicle behavior data, pedestrian dynamic data, and user response data, and analyzes key indicators.

2. The simulated driving vision training system for people with low vision according to claim 1, characterized in that: The vehicle behavior data includes real-time speed, location and collision information.

3. The simulated driving vision training system for people with low vision according to claim 1, characterized in that: The pedestrian dynamic data includes initial position, movement speed, movement trajectory and interaction events.

4. The simulated driving vision training system for people with low vision according to claim 1, characterized in that: The data storage and analysis module stores all data during the driving process at a frequency of 30 Hz.

5. The simulated driving vision training system for people with low vision according to claim 1, characterized in that: The events that attract attention include oncoming vehicles, following vehicles, vehicles coming from intersections, and vehicles coming from behind.

6. The simulated driving vision training system for people with low vision according to claim 1, characterized in that: The response data includes steering wheel rotation, button operation and pedal data.

Images