EP4356363A1

EP4356363A1 - System and method for increasing traffic safety

Info

Publication number: EP4356363A1
Application number: EP21745781.1A
Authority: EP
Inventors: Mart SUURKASK; Werner RUUL
Original assignee: Bercman Technologies AS
Current assignee: Bercman Technologies AS
Priority date: 2021-06-15
Filing date: 2021-07-15
Publication date: 2024-04-24
Also published as: WO2022263008A1

Abstract

A system for increasing traffic safety comprising a monitoring component that is configured to monitor traffic participants and an analyzing component that is configured to predict a range of trajectories of the traffic participants. The analyzing component predicts the range of trajectories with a frequency of 1 per second, preferably or more than 10 per second. The range of trajectories comprises a value representing the likelihood of one or more trajectories, wherein the analyzing component comprises trained software that has been trained with respect to the likelihood of prediction of the trajectories and/or the precision of the trajectories of the traffic participants.

Description

System and method for increasing traffic safety

Field

The present invention relates to a system and a method for increasing the safety of traffic. It is particularly directed to avoid accidents in traffic and/or to minimize any injuries.

Introduction

In general, globally, vulnerable road users - VRUs (e.g., pedestrians and cyclists) account for more than 50% of all traffic deaths. Statistics show that more than one-third of VRU involved accidents take place on uncontrolled (zebra) crossings. However, it should be noted that these fatalities represent only a part of the traffic safety problem. These accidents also result in non-fatal injuries, some requiring long-term care and rehabilitation.

Considering that over 90% of all traffic accidents are attributable to human error, educating road users of potential road hazards are not enough to reach the ambitious traffic safety goals set by most developed countries. The only way to solve this issue is through the approach to avoid any accidents. This implies that traffic systems must be designed with human error in mind, and they should be forgiving, and crashes should not result in death or severe injury.

Nonetheless, achieving a total avoidance of accidents for all road user groups has proven difficult as there are currently no cost-effective solutions that can significantly improve traffic safety for VRUs in dynamic urban environments.

US 9928734 B2 is directed to vehicle-to-pedestrian information systems that use directional sound transmission on autonomous vehicles are disclosed. A cloud computing system manages messages for transmission to pedestrians via autonomous vehicles having directional speakers. The cloud computing system identifies pedestrians and identifies messages for the pedestrians. Pedestrians may be known and authenticated to the cloud computing system or may be unknown. The cloud computing system maintains profiles for known pedestrians and transmits messages to vehicles based on the profiles. The cloud computing system keeps track of the location of vehicles and causes the vehicles to use directional speakers to transmit messages to the pedestrians based on the relative positions of the vehicles and the pedestrians. KR 101806470 B1 relates to a pedestrian protecting system using a beacon signal capable of sensing a vehicle approaching a crosswalk by recognizing the beacon signal transmitted from a beacon device installed in the vehicle; informing approach of the vehicle to a pedestrian or a waiting person on a sidewalk; and preventing an accident in the crosswalk by enabling a driver of the vehicle to confirm the pedestrian or the waiting person on the sidewalk.

JP 2015197785 A is directed to a pedestrian support system and a pedestrian support program capable of providing walking safety support for pedestrians including visually impaired persons. Described is a pedestrian terminal that can be carried by a pedestrian and an automobile terminal provided in a vehicle having a steering, and the position, speed, direction of the vehicle of the automobile terminal 200, and steering of the steering. It touches also the determining of the presence or absence of crossing between the predicted progress area of the vehicle in one time zone calculated from the corner and the predicted progress area of the pedestrian in the one time zone calculated from the information of the pedestrian. The pedestrian terminal is notified of the presence of the vehicle based on the determination result of the presence or absence of crossing. Thereby, the collision with a pedestrian and a vehicle can be prevented beforehand.

US 201800906595 A1 relates to traffic signal control systems and methods in accordance with various embodiments of the invention. One embodiment includes: at least one image sensor mounted with a bird's eye view of an intersection; memory containing a traffic optimization application and classifier parameters for a plurality of classifiers, where each classifier is configured to detect a different class of object, and a processing system. In addition, the traffic optimization application directs the processing system to: capture image data using the at least one image sensor; search for pedestrians and vehicles visible within the captured image data by performing a plurality of classification processes based upon the classifier parameters for each of the plurality of classifiers; determine modifications to the traffic signal phasing based upon detection of at least one of a pedestrian or a vehicle; and send traffic signal phasing instructions to a traffic controller directing modification to the traffic signal phasing.

US 9368028 B2 concerns techniques for ability enhancement. Some embodiments provide an ability enhancement facilitator system ("AEFS") configured to enhance a user's ability to operate or function in a transportation-related context as a pedestrian or a vehicle operator. In one embodiment, the AEFS is configured to perform vehicular threat detection based on information received at a road-based device, such as a sensor or processor that is deployed at the side of a road. An example AEFS receives, at a road-based device, information about a first vehicle that is proximate to the road-based device. The AEFS analyzes the received information to determine threat information, such as that the vehicle may collide with the user. The AEFS then informs the user of the determined threat information, such as by transmitting a warning to a wearable device configured to present the warning to the user.

Summary

In light ofthe above, it is an object of the present invention to overcome or at least alleviate the shortcomings of the prior art.

The invention describes a system for increasing traffic safety that can comprise a monitoring component that may be configured to monitor traffic participants; and an analyzing component that can be configured to predict a range of trajectories of the traffic participants.

Monitoring components may comprise any kind of sensor, such as electronical and/or optical and/or acoustic sensors that detect participants in a traffic environment, like radar- or laser-sensors, cameras or video-devices. Monitoring components may also comprise magnetic field sensors, photo-electric barriers that send their data to an analyzing component. Also, human intervention, like, but not limited to, gatekeeper operated lights may control or contribute information to an analyzing component.

Traffic participants may be considered to comprise vehicles of any kind, like cars, trucks, motorcycles, bicycles, scooters, but also pedestrians and others.

An analyzing component may be addressed as a computer, be it a local device or a computing instance in the cloud. The analyzing component may further comprise elements of artificial intelligence, machine learning algorithms, neuronal network device.

Where a range of trajectories is referenced, predictable or predicted paths are meant that a traffic participant can probably take. While a car or a truck usually may comprise steady trajectories, like straight lines, curves or a combination thereof, a pedestrian may be more difficult to predict. A pedestrian may decide to stop on the spot, decide to proceed in another direction or, more general, his or her trajectory is hard to predict. Therefore, the term "range of trajectories" is used because possible, rather presumable, even more likely motion patterns and their ranges or tolerances may be taken into account by the analyzing component. In other words, a range of trajectories comprises a plurality of trajectories that appear possible. Their likelihood of a later realization may be comprised in this range of trajectories.

The range of trajectories may be estimated at a high frequency, i.e., for instance, at least 1 time per second or more often, like 10 times per second or faster. With more than 15 time, the analyzing component determining the location and the estimated trajectories can be taken into account and lead to precautionary measures.

The trajectory of a traffic participant may comprise a number of ranges, named the range of trajectories. While it is unlikely, but possible, that a car or another traffic participant may stop and even reverse its prior direction, this is still possible. The analyzing component may determine the likelihood of a certain trajectory based on statistically determined trajectories. The trajectories may also derive from prior learning process using methods of artificial learning. A bunch of trajectories may be determined and applied with probability values that can be used for decisions to be made by the analyzing component.

The analyzing component may comprise a software that has been trained before installation; however, even during every-day use, the analyzing component may also learn based on trajectories provided by the sensor elements. Taking the trajectories of at least one traffic participant and comparing them to prior determinations of trajectories of other traffic participants results in a set of experiences how trajectories can be assumed in the future. This enables the analyzing component to apply probability factors to each possible trajectory. Very unprobeable possibilities may be neglected or considered at a lower value. By applying training (self-learning) algorithms enables the analyzing component to increase the precession of a prediction or forecast.

A huge number of neural network methods are known by the person skilled in the art. One or a combination of neural networks may be applied, like, but not limited to, a Markov Chain, a Kohonen Network, a Neural Turing Machine, a Hopefield Network, a Boltzmann Machine, a TensorRT engine or a Deep Belief Network.

Other deep neural networks may be comprised like the already mentioned Deep Belief Network, but also a Deep Convolutional Network, a Deep Deconvolutional Network, also inverted, double networks like a Generative Adversarial Network,

The system for increasing traffic safety may comprise an internal and an external storage or server where data received from any of the sensors can be fed in, calculations can be accomplished and/or where the actual computations are performed. Alternatively, or additionally, parts of the computational work performed by a computer may be exported to a remote instance. The software under training or being trained may be centralized and/or distributed to one or a number of on-site analyzing components.

The system may further comprise a remote storage or server for storing data and/or software for training the software. Also, or alternatively, the system may comprise a remote storage or server for storing data and/or software for the analyzing component.

Further, the system may comprise a remote storage or server for storing data and/or software for the analyzing component.

The system for increasing traffic safety may comprise one or more cameras to capture traffic images of a traffic location. The camera or the cameras may be of the type video camera, stereo camera, both covering visible light wavelengths, but also infrared or any range of wavelengths appropriate for the current location or situation.

The sensor may comprise at least one camera, preferably a plurality of cameras that may capture a far overview of the location of installation. A wide angle may capture a traffic location up to 360°.

The software may be configured to analyze static or dynamic images; the trained software may analyze images taken by the at least one sensor and may deliver its resulted data to the analyzing component for further processing. Further, the trained software may be able to detect accidents or unusual behavior of a traffic participant.

Further, the system for increasing traffic safety may comprise a warning component sending out a warning signal, like a horn, a bell, a flashlight, floodlight, traffic signals and/or an electronic device to wirelessly warn traffic participants in case of a possible overlap of trajectories. This may be done not only having analyzed static trajectories but would also be enabled to determine whether crossing or merging trajectories may independently occur at the same time. The intersecting or merging trajectories may be analyzed taking probabilities into account, that is, not every possible trajectory of a traffic participant is likely. In this case, the warning signal may be initiated if a defined threshold of probabilities is determined.

The warning signal may comprise an optical, an acoustical and/or electronical signal sent to either a specific traffic participant and/or to a plurality of traffic participants. Further, light and/or sound signals may be sent to an appropriate device, like a LED light, a flash light, traffic signal or any other installed device at or in the vicinity the location of interest.

In an embodiment, a flash light may be activated by a traffic participant, for instance, but not limited to, a pedestrian.

A warning signal may also comprise a backlighted traffic signal that is installed, for instance, at a zebra path to be used by pedestrians. Also, regular traffic lights may be triggered by human activity.

Bells, loudspeakers, horns or similar acoustic may also be comprised by the system. Also, an optical may be comprised. Recently, simulated 3D zebra crossings have been introduced. This illusion (or simulation) may also be used to warn traffic participants.

The system for increasing traffic safety may comprise a sound detecting system, like a microphone would constitute. The signal delivered by the sound detecting system may be fed into the system to capture traffic information.

The system may also comprise an accident information component like eCalls that can be adapted to alert an emergency institution and/or an ambulance. This may even be accomplished in realtime or at least quasi-realtime.

The system may usually be fixedly installed at a traffic location, but may also be mobile to cover dangerous situations that may occasionally occur. For this, a GPS (global positioning) or hyperboloid position finding component may be fitted.

The alert to an emergency institution may also be provided with a communication provision that may be of a wireless type A wireless communication may be accomplished making use of mobile phone network or mobile signaling A mobile phone network may be one or a combination of 4G or 5G type.

The system for increasing traffic safety may comprise a communication component configured to communicate with communication-ready traffic participants and/or vehicles and/or handheld devices. This can e done by a point-to-point communication or of a broadcast type.

The system may comprise a smart pedestrian crosswalk component (SPC) The monitoring component can comprise a plurality of sensors configured to monitor the traffic participants. The sensors may be configured to be self-calibrate their readouts.

Also, weather conditions may be kept track of to estimate braking distances or steering conditions and take them into account when predicting probable trajectories.

The weather conditions may comprise an environmental temperature sensor, a road temperature sensor, a windspeed sensor, a wind direction sensor, a humidity detector and/or a visibility detector.

The system or parts of the system may further comprise a range of power supplies that may comprise reloadable batteries (accumulators), single-use batteries or a fixed grid power supply. An accumulator may be charged by solar panels, wind-power and/or mechanical power.

A range of trajectories is intended to comprise a range of a prospective way and time of progress of a traffic participant. This range can be re-calculated frequently in order to limit and/or converge the possible ranges. The range of trajectories can comprise a dataset representing it and can comprise a value representing the likelihood of one or more trajectories.

Methods for increasing traffic safety may comprise monitoring of traffic participants by a monitoring component. The method further may comprise a prediction of a range of trajectories of the traffic participants by an analyzing component.

Traffic participants can comprise trucks, cars, vans, motorcycles, bicycles, scooters, but also pedestrians. The method can make use of an analyzing component.

The analysis can be performed inside the system without internet connection. This may be called a local intelligence - device can make decisions on its own.

The method may comprise the analysis being performed locally and/or without instructions from an external instance.

Also, the method may derive the results of the analysis to decide to carry out measures, like initiating a warning measure, an alarm, decide on switching on or off light signals.

A monitoring component may comprise any kind of sensor, such as electronical and/or optical and/or acoustic sensors that detect participants in a traffic environment, like radar- or laser-sensors, cameras or video-devices. Monitoring components may also comprise magnetic field sensors, photo-electric barriers that send their data to an analyzing component. Also, human intervention, like, but not limited to, gatekeeper operated lights may control or contribute information to an analyzing component. Also, an induction loop, a photoelectric barrier may serve as a monitoring component.

Where a range of trajectories is referenced, predictable or predicted paths are meant that a traffic participant can take. While a car or a truck usually may comprise steady trajectories, like straight lines, curves or a combination thereof, a pedestrian may be more difficult to predict. A pedestrian may decide to stop on the spot, decide to proceed in another direction or, more general, his or her trajectory is hard to predict. Therefore, the term "range of trajectories" is used because possible, rather presumable, even more likely motion patterns and their ranges or tolerances may be taken into account by the analyzing component. In other words, a range of trajectories comprises a plurality of trajectories that appear possible. Their likelihood of a later realization may be comprised in this range of trajectories.

A number of analytic methods may be comprised. As examples, a neural network or a deep neural network may be comprised, also, a combination of different methods of nor systematic algorithms may be applied, more than one deciding instance may further serve the method to estimate or predict a situation and/or trajectories of traffic participants.

The method may further comprise a remote storage for storing data and/or software under training or being trained for the analyzing component and/or a remote storage for storing data and/or software for training the software. Further, the method can comprise a remote storage for storing data and/or software for the analyzing component and/or can comprise the step of storing data and/or software for the analyzing component in a remote storage or server.

The Method can further comprise at least one, preferably two or more cameras and capturing traffic images of a traffic location with the camera(s). Where a camera is referenced, other image creating sensors may be addressed, like radar-, lidar- and ultrasonic sensors or a combination thereof. The sensor or the plurality of sensors may comprise a capturing range of up to 360°, which means a full overview. Other sensors of the same or differing sensing principle may focus on reduced angles, be it because of constructional obstacles or the need to cover a certain detail of a traffic situation in more accuracy.

The Method can further comprise a trained software for analyzing images, wherein the trained software may be configured to detect an accident or another unusual traffic situation. In some cases, certain predefined measures can be provided for security reasons.

In normal operation environment, but more important in unusual or emergency situations a warning component may be configured to indicate warning(s) like blowing a sound signal. Other than a loudspeaker sound, also a bell, a whistle, or optical warnings may be configured, such as flashlights, ordinary traffic lights, extended lighting (for instance with spotlights), with LED lights may be initiated. Also, wireless warnings can be transmitted to traffic participants that are equipped with appropriate devices.

The warning signal may also be initiated in case their range of trajectories overlap.

Further to automated operation, a traffic participant, for instance a pedestrian, may activate his or her wish to pass a zebra path by pressing an appropriate signal means.

Also, a traffic sign can be backlighted or focused by a spotlight if a situation may have been analyzed by the analyzing component.

The method may further comprise listening to the surroundings by the use of a microphone. Unusual sounds may indicate an unusual or unwanted situation which can be analyzed by the analyzing component.

If the analyzing component estimates an unusual, a dangerous situation, like an accident would comprise, the analyzing component or the warning component may initiate a wired or a wirelessly transmitted notice to an emergency institution, like an ambulance. Such systems are established, like eCall network. The possibility to use currently available communication means may be obvious. The method may further comprise detecting the current location by making use of hyperbolic position finding systems, like GPS, Galileo, Glonass, Baidoo or others.

The method may further comprise the communication via a communication component to traffic participants that may have a counterpart communication providence. Other than in vehicles, also handheld devices may be addressed. This may be accomplished by dialogue- based communication, but also broadcast communication may be comprised, or a combination thereof. A smart pedestrian crosswalk component may be comprised using the method.

The method further comprises the ability to receive sensor data from installed sensors, but also data deriving from cars that broadcast their detections to the surrounding traffic participants. Even the intention of a vehicle can be broadcast, for instance, if the driver ticks the indicator lever, the method may detect this and take this information into account when determining the probability of trajectories. It may be more probable that a car would turn right at a crossing, if the car uses the right indicators (or left, of course, or straight, if no indicator is set). Further, or alternatively, also the motion of the steering wheel may be used to enhance the probability for a trajectory forecast.

The method may further comprise the collection of a multifold of sensor readouts by the monitoring component to keep track of the traffic participants by closely monitoring each of the traffic participants; this may be done more than 1 time a second, even more than 15 times per second or more often if needed.

The method may comprise sensors that are enabled to comprise self-calibrating corrections.

The method may include weather conditions, that can comprise an environmental temperature sensor, a road temperature sensor, a windspeed sensor, a wind direction sensor, a humidity detector and/or a visibility detector.

The method may comprise a step of charging by a range of power supplies that may comprise reloadable batteries (accumulators), single-use batteries or a fixed grid power supply. An accumulator may be charged by solar panels, wind-power and/or mechanical power.

Below, system embodiments will be discussed. These embodiments are abbreviated by the letter "S" followed by a number. Whenever reference is herein made to "system embodiments", these embodiments are meant. 51. System for increasing traffic safety comprising: a. a monitoring component (3) that is configured to monitor traffic participants (10-13); and b. an analyzing component (1) that is configured to predict a range of trajectories (10a-13a) of the traffic participants.

52. System according to the preceding system embodiment wherein the analyzing component is configured to predict the range of trajectories with a frequency of at least 1 per second, preferably at least 10 per second and more preferably at least 15 per second.

53. System according to any of the preceding system embodiments wherein the range of trajectories (10a-13a) comprise a value representing the likelihood of one or more trajectories.

54. System according to any of the preceding system embodiments the analyzing component (1) comprises trained software that has been trained with respect to the likelihood of prediction of the trajectories and/or the precision of the trajectories of the traffic participants.

55. System according to any of the preceding system embodiments wherein the analyzing component (1) comprises a neural network.

56. System according to the preceding system embodiment wherein the analyzing component (1) comprises a deep neural network.

57. System according to the preceding system embodiment, wherein the deep neural network is a TensorRT engine type.

58. System according to any of the preceding system embodiments further comprising a remote storage or server (2) for storing data and/or software under training or being trained for the analyzing component (1).

59. System according to any of the preceding system embodiments further comprising a remote storage or server (2) for storing data and/or software for training the software. 510. System according to any of the preceding system embodiments further comprising a remote storage or server (2) for storing data and/or software for the analyzing component (1)

511. System according to any of the preceding system embodiments further comprising a remote storage or server (2) for storing data and/or software for the analyzing component (1)

512. System according to any of the preceding system embodiments further comprising at least one, preferably two or more cameras to capture traffic images of a traffic location.

513. System according to any of the preceding system embodiments further comprising at least one, preferably two or more cameras (3) to capture 360° traffic images of a traffic location.

514. System according the any one of the two preceding system embodiments further comprising a trained software for analyzing images.

515. System according the any one of the three preceding system embodiments further comprising a trained software for analyzing images and for detecting accidents.

516. System according to any of the preceding system embodiments further comprising a warning component (5) that is configured to warn traffic participants (10-13) in case their range of trajectories overlap.

517. System according to any of the preceding system embodiments further comprising LED signal lights.

518. System according to any of the preceding system embodiments further comprising a pedestrian-activated flash.

519. System according to any of the preceding system embodiments further comprising a backlit traffic sign (5).

520. System according to any of the preceding system embodiments further comprising a loudspeaker.

521. System according to any of the preceding system embodiments further comprising a microphone configured for capturing acoustic traffic information. 522. System according to any of the preceding system embodiments further comprising an accident information component (eCalls) that is adapted to alert an emergency institution (20) and/or an ambulance (21).

523. System according to any of the preceding system embodiments further comprising an accident information component (eCalls) comprising a GPS (global positioning) module.

524. System according to any of the preceding system embodiments further comprising an accident information component (eCalls) comprising a wireless communication module, such as 4G, 5G and/or 6G.

525. System according to any of the preceding system embodiments further comprising a communication component configured to communicate with communication-ready traffic participants and/or vehicles and/or handheld devices.

526. System according to any of the preceding system embodiments further comprising a smart pedestrian crosswalk component (SPC).

527. System according to any of the preceding system embodiments wherein the monitoring component comprises a plurality of sensors configured to monitor the traffic participants.

528. System according to the preceding system embodiment wherein the sensors are configured for self-calibration.

529. System according to any of the preceding system embodiments further comprising a weather sensor.

530. System according to any of the preceding system embodiments further comprising an environment temperature sensor.

531. System according to any of the preceding system embodiments further comprising a road temperature sensor.

532. System according to any of the preceding system embodiments further comprising a solar panel and a battery for delivering the energy for the system or parts thereof.

Below, method embodiments will be discussed. These embodiments are abbreviated by the letter "M" followed by a number. Whenever reference is herein made to "method embodiments", these embodiments are meant. Ml. Method for increasing traffic safety comprising: a. monitoring traffic participants (10-13) by a monitoring component (3); and b. predicting a range of trajectories (10a-13a) of the traffic participants by an analyzing component (1).

M2. Method according to the preceding method embodiment wherein the analyzing component is configured to predict the range of trajectories with a frequency of at least 1 per second, preferably at least 10 per second and more preferably at least 15 per second.

M3. Method according to any of the preceding method embodiments, wherein an analysis is performed locally and/or without instructions from an external instance.

M4. Method according to any of the preceding method embodiments, wherein a decision on further measures is based on the analysis.

M5. Method according to any of the preceding method embodiments wherein the range of trajectories (10a-13a) comprises a value representing the likelihood of one or more trajectories.

M6. Method according to the preceding method embodiment further with the step of training software of the analyzing component (1) with respect to the likelihood of prediction of the trajectories and/or the precision of the trajectories of the traffic participants.

M7. Method according to any of the preceding method embodiments wherein the analyzing component (1) comprises a neural network.

M8. Method according to the preceding method embodiment wherein the analyzing component (1) comprises a deep neural network.

M9. Method according to any of the preceding method embodiments further comprising a remote storage (2) for storing data and/or software under training or being trained for the analyzing component (1).

M10. Method according to any of the preceding method embodiments further comprising a remote storage (2) for storing data and/or software for training the software. Mi l. Method according to any of the preceding method embodiments further comprising a remote storage (2) for storing data and/or software for the analyzing component

(1). 12. Method according to any of the preceding method embodiments further comprising the step of storing data and/or software for the analyzing component (1) in a remote storage or server (2).

M13. Method according to any of the preceding method embodiments further comprising at least one, preferably two or more cameras (3) and capturing traffic images of a traffic location with the camera(s).

M14. Method according to any of the preceding method embodiments further comprising at least one, preferably two or more cameras (3) to capture an area of interest traffic images of a traffic location up to 360° coverage.

M15. Method according the any one of the two preceding method embodiments further comprising a trained software for analyzing images. 16. Method according the any one of the three preceding method embodiments further comprising a trained software for analyzing images and for detecting accidents.

M17. Method according to any of the preceding method embodiments further comprising a warning component (5) that is configured to warn traffic participants (10-13) in case their range of trajectories overlap.

M18. Method according to any of the preceding method embodiments further comprising LED signal lights.

M19. Method according to any of the preceding method embodiments further comprising a pedestrian-activated flash.

M20. Method according to any of the preceding method embodiments further comprising a backlit traffic sign (5).

M21. Method according to any of the preceding method embodiments further comprising a loudspeaker.

M22. Method according to any of the preceding method embodiments further comprising a microphone configured for capturing acoustic traffic information. M23. Method according to any of the preceding method embodiments further comprising an accident information component (eCalls) that is adapted to alert an emergency institution (20) and/or an ambulance (21)

M24 Method according to any of the preceding method embodiments further comprising an accident information component (eCalls) that is adapted to alert an emergency institution and/or an ambulance in realtime.

M25. Method according to any of the preceding method embodiments further comprising an accident information component (eCalls) comprising a GPS (global positioning) module.

M26. Method according to any of the preceding method embodiments further comprising an accident information component (eCalls) comprising a wireless communication module, such as 4G, 5G and/or 6G.

M27. Method according to any of the preceding method embodiments further comprising a communication component configured to communicate with communication-ready traffic participants and/or vehicles and/or handheld devices

M28. Method according to any of the preceding method embodiments further comprising a smart pedestrian crosswalk component (SPC).

M29. Method according to any of the preceding method embodiments wherein the monitoring component comprises a plurality of sensors configured to monitor the traffic participants

M30 Method according to the preceding method embodiment wherein the sensors are configured for self-calibration.

M31. Method according to any of the preceding method embodiments further comprising a weather sensor.

M32. Method according to any of the preceding method embodiments further comprising an environment temperature sensor.

M33. Method according to any of the preceding method embodiments further comprising a road temperature sensor.

M34. Method according to any of the preceding method embodiments further comprising a solar panel and a battery for delivering the energy for the system or parts thereof. Below, use embodiments will be discussed. These embodiments are abbreviated by the letter "U" followed by a number. Whenever reference is herein made to "use embodiments", these embodiments are meant.

Ul. Use of the system according to any of the preceding system embodiments for increasing traffic safety of by applying the system according to any of the preceding system embodiments.

U2. Use of the method according to any of the preceding method embodiments for increasing traffic safety of by carrying out the method according to any of the preceding method embodiments.

The present invention will now be described with reference to the accompanying drawings, which illustrate embodiments of the invention. These embodiments should only exemplify, but not limit, the present invention.

Brief Figure Description

Fig. 1 schematically exemplifies a flowchart in accordance with an embodiment according to the invention.

Fig. 2 a principal view onto a traffic section with a system and a respective method according to the present invention.

Fig. 3 depicts a symbolized configuration of a monitoring component.

Fig. 4 depicts an embodiment of a detection architecture.

Fig. 5 depicts the pre-training of a Trained Neural Network.

Description of preferred embodiments as exemplified in the figures

It is noted that not all the drawings carry all the reference signs. Instead, in some of the drawings, some of the reference signs have been omitted for sake of brevity and simplicity of illustration. Embodiments of the present invention will now be described with reference to the accompanying drawings.

Fig. 1 provides a schematic of a computing device 100. The computing device 100 may comprise a computing unit 35, a first data storage unit 30A, a second data storage unit 30B and a third data storage unit 30C. The computing device 100 can be a single computing device or an assembly of computing devices. The computing device 100 can be locally arranged or remotely, such as a cloud solution.

On the different data storage units 30 the different data can be stored. Additional data storages can be also provided and/or the ones mentioned before can be combined at least in part.

The computing unit 35 can access the first data storage unit 30A, the second data storage unit 30B and the third data storage unit 30C through the internal communication channel 160, which can comprise a bus connection 160.

The computing unit 30 may be single processor or a plurality of processors, and may be, but not limited to, a CPU (central processing unit), GPU (graphical processing unit), DSP (digital signal processor), APU (accelerator processing unit), ASIC (application-specific integrated circuit), ASIP (application-specific instruction-set processor) or FPGA (field programable gate array). The first data storage unit 30A may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM).

The second data storage unit 30B may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM). The third data storage unit 30C may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM).

It should be understood that generally, the first data storage unit 30A (also referred to as encryption key storage unit 30A), the second data storage unit 30B (also referred to as data share storage unit 30B), and the third data storage unit 30C (also referred to as decryption key storage unit 30C) can also be part of the same memory. That is, only one general data storage unit 30 per device may be provided, which may be configured to store the respective encryption key (such that the section of the data storage unit 30 storing the encryption key may be the encryption key storage unit 30A), the respective data element share (such that the section of the data storage unit 30 storing the data element share may be the data share storage unit 30B), and the respective decryption key (such that the section of the data storage unit 30 storing the decryption key may be the decryption key storage unit 30A).

In some embodiments, the third data storage unit 30C can be a secure memory device 30C, such as, a self-encrypted memory, hardware-based full disk encryption memory and the like which can automatically encrypt all of the stored data. The data can be decrypted from the memory component only upon successful authentication of the party requiring to access the third data storage unit 30C, wherein the party can be a user, computing device, processing unit and the like. In some embodiments, the third data storage unit 30C can only be connected to the computing unit 35 and the computing unit 35 can be configured to never output the data received from the third data storage unit 30C. This can ensure a secure storing and handling of the encryption key (i.e., private key) stored in the third data storage unit 30C.

In some embodiments, the second data storage unit 30B may not be provided but instead the computing device 100 can be configured to receive a corresponding encrypted share from the database 60. In some embodiments, the computing device 100 may comprise the second data storage unit 30B and can be configured to receive a corresponding encrypted share from the database 60.

The computing device 100 may comprise a further memory component 140 which may be singular or plural, and may be, but not limited to, a volatile or non-volatile memory, such as a random-access memory (RAM), Dynamic RAM (DRAM), Synchronous Dynamic RAM (SDRAM), static RAM (SRAM), Flash Memory, Magneto-resistive RAM (MRAM), Ferroelectric RAM (F-RAM), or Parameter RAM (P-RAM). The memory component 140 may also be connected with the other components of the computing device 100 (such as the computing component 35) through the internal communication channel 160.

Further the computing device 100 may comprise an external communication component 130. The external communication component 130 can be configured to facilitate sending and/or receiving data to/from an external device (e.g., backup device, recovery device, database). The external communication component 130 may comprise an antenna (e.g., WIFI antenna, NFC antenna, 2G/3G/4G/5G antenna and the like), USB port/plug, LAN port/plug, contact pads offering electrical connectivity and the like. The external communication component 130 can send and/or receive data based on a communication protocol which can comprise instructions for sending and/or receiving data. Said instructions can be stored in the memory component 140 and can be executed by the computing unit 35 and/or external communication component 130. The external communication component 130 can be connected to the internal communication component 160. Thus, data received by the external communication component 130 can be provided to the memory component 140, computing unit 35, first data storage unit 30A and/or second data storage unit 30B and/or third data storage unit 30C. Similarly, data stored on the memory component 140, first data storage unit 30A and/or second data storage unit 30B and/or third data storage unit 30C and/or data generated by the computing unit 35 can be provided to the external communication component 130 for being transmitted to an external device.

In addition, the computing device 100 may comprise an input user interface 110 which can allow the user of the computing device 100 to provide at least one input (e.g., instruction) to the computing device 100. For example, the input user interface 110 may comprise a button, keyboard, trackpad, mouse, touchscreen, joystick and the like.

Additionally, still, the computing device 100 may comprise an output user interface 120 which can allow the computing device 100 to provide indications to the user. For example, the output user interface 110 may be a LED, a display, a speaker and the like.

The output and the input user interface 100 may also be connected through the internal communication component 160 with the internal component of the device 100.

The processor may be singular or plural, and may be, but not limited to, a CPU, GPU, DSP, APU, or FPGA. The memory may be singular or plural, and may be, but not limited to, being volatile or non-volatile, such an SDRAM, DRAM, SRAM, Flash Memory, MRAM, F-RAM, or P-RAM.

The data processing device can comprise means of data processing, such as, processor units, hardware accelerators and/or microcontrollers. The data processing device 20 can comprise memory components, such as, main memory (e.g., RAM), cache memory (e.g., SRAM) and/or secondary memory (e.g., HDD, SDD). The data processing device can comprise busses configured to facilitate data exchange between components of the data processing device, such as, the communication between the memory components and the processing components. The data processing device can comprise network interface cards that can be configured to connect the data processing device to a network, such as, to the Internet. The data processing device can comprise user interfaces, such as: ^■ output user interface, such as:

• screens or monitors configured to display visual data (e.g., displaying graphical user interfaces of the questionnaire to the user),

• speakers configured to communicate audio data (e.g., playing audio data to the user),

^■ input user interface, such as:

• camera configured to capture visual data (e.g., capturing images and/or videos of the user),

• microphone configured to capture audio data (e.g., recording audio from the user),

• keyboard configured to allow the insertion of text and/or other keyboard commands (e.g. allowing the user to enter text data and/or other keyboard commands by having the user type on the keyboard) and/or trackpad, mouse, touchscreen, joystick - configured to facilitate the navigation through different graphical user interfaces of the questionnaire.

The data processing device can be a processing unit configured to carry out instructions of a program. The data processing device can be a system-on-chip comprising processing units, memory components and busses. The data processing device can be a personal computer, a laptop, a pocket computer, a smartphone, a tablet computer. The data processing device can be a server, either local and/or remote. The data processing device can be a processing unit or a system-on-chip that can be interfaced with a personal computer, a laptop, a pocket computer, a smartphone, a tablet computer and/or user interface (such as the upper-mentioned user interfaces).

Fig. 2 exemplifies a street crossing with a traffic situation that can be significantly improved by the present invention. Shown is a street crossing with two zebra crossings for pedestrians. One pedestrian 13 and a plurality of other traffic participants 10-12 are shown, namely a bicycle rider 10 and two cars 11, 12.

With a camera 3 the crossing is monitored. This constitutes just an example; more cameras and other sensors can be provided and are not shown for the sake of ease. The camera 3 is capturing the situation and delivers this information to an analyzing component 1. The camera 3 and sensors (not shown) can communicate with the analyzing component wireless and/or hardwired. Also, other components, such as a traffic sign 5 with or without other sensors, can also communicated with the analyzing component 1. A remote storage or server 2 can be remotely located and collect data from the analyzing component 1 and/or can deliver trained software to the analyzing component 1. The latter implies that the software can be trained remotely by collecting data from different sites and then train the software for the analyzing component(s) 1.

Shown are also the trajectories 10a-13a of traffic participants as projected or predicted by the analyzing component 1. As an aside, also the analyzing component 1 could be located remotely, such as in the cloud. Anyhow, the range of trajectories means all or the most probably potential paths or ways the traffic participants will take. This can be calculated with a high frequency in order to renew the prediction and make it more and more precise.

In the examples shown, the range of trajectories 10a for the bicycle 10 is rather broad as there appears no indication where the rider of the bicycle will go. The rider could go straight or make a left or a right turn. As the rider is close to the crossing and as he hasn't provided any signal, the way straight is the most probable one. This may be reflected by the range or dataset representing the range. The car 11 has provided a sign to take a right turn so that the respective range or dataset comprises a high probability for the right-hand turn and low or very low or zero probabilities for the other directions. For the second car 12 the relevant trajectory for the pedestrian 13 is the straight one. For other pedestrians (not shown) standing at other points of the crossing it may be different.

An emergency institution 22 with an emergency vehicle 21 is also exemplified that can be connected with the analyzing component 1 so that this component can trigger an alarm so that the emergency vehicle can approach the location of an accident very quickly. This can be a life saving measure.

Fig. 3 depicts an embodiment of the invention. A monitoring component may capture images through a video camera that may be connected by an IP link, the streams can be encoded using series of signal processing techniques (see figure 1), fed into a neural network through a TensorRT engine (VRU detection module, a high-performance deep learning inference) with pre-trained weights. The output of the TensorRT is detection parameters including; IDs and coordinates which can be used to draw bounding boxes around detected pedestrians or other traffic participants. Furthermore, the detection module's output can be fed into the pedestrian tracking algorithm which may estimate the pedestrian statistics; such as pedestrian speed, pedestrian count, etc. The General Overview of the detection module is schematically presented in figure 4 and TensortRT® architecture is presented in figure 5. Camera can capture frames of images and the stream can then be compressed into the desired video encoding format, such as MPEG (but can be changed). The audio data is muted out in this embodiment. Thereafter the video data is passed into Real-Time Streaming Protocol (RTSP). The protocol can be used for establishing and controlling media sessions between the camera and neural network engine that can use the stream as input data to perform detection operations. The output stream may be handled by Internet Protocol, for instance, version 6 (IPv6) which is the communications protocol that provides identification on networks and routes traffic across the Internet.

The camera in this embodiment can be replaced (or added by) radar, lidar detection devices; the corresponding signals may be handled in a similar procedure as disclosed but adapted to the features of the detection devices.

Fig. 4 depicts a detection architecture. The whole streaming procedure as discussed in figure 3 is represented here in this figure as the "webcam camera". A processor may parse any data format, preferably a universal file formatted dataset (UFF format) and ...

Other suitable file formats are, but not limited to, NCHW and/or Cv: :mat.

The camera records images and can feed it into tensortRT and the tensortRT can encode it into its preferred file type (for instance, NVCUVID format) and may accelerates the inference process (detection). The inference process can output the coordinates of the detected objects (bounding boxes) and it also can annotate the objects (identify what type of object). From this analysis, this information can be used to design the algorithm behind the detection, tracking and counting.

Fig. 5 depicts an exemplary Nvidia® TensortRT® optimizer installed in a SDK (system development kit) architecture.

By application of Reduced mixed-precision a maximum throughput can be achieved by quantizing models to INT8 while preserving accuracy.

Further, a layer and tensor merge, thereby optimizes use of GPU (graphic processor unit) memory and bandwidth by fusing nodes in a kernel.

The kernel can be auto-tuned by selecting best data layers and algorithms based on the target GPU platform. The platform applies Dynamic Tensor memory management; it minimizes memory footprint and reuses memory for tensors efficiently.

Further, by making use of multi-stream execution, a scalable use is made for a design to process multiple input streams in parallel.

During operational time, the neural network optimizes steps with dynamically generated kernels.

In an example to disclose to the person skilled in the art, a typical scenery is illustrated. TensorRT is a software created by NVidia to help developers accelerate performance on NVidia GPUs. TensorRT maximizes throughput by quantizing models to INT8 while preserving accuracy, and automatically selects best data layers and algorithms that are optimized for the target GPU platform. This may be accomplished, but is not limited to, by NVidia jetson nano.

Unlike many other GPU accelerated boards, Nvidia jetson nano (with cuda) has been made such that it can only processing certain types of image or videos. Common image format from cameras such as MPEGs and H.264 video formats will be converted to NVCUVID format through the VP2(VP2 is a dedicated video-decode engine on NVIDIA GPUs) platform onboard.

Embodiments

While the disclosure has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and non-restrictive; the disclosure is thus not limited to the disclosed embodiments. Variations to the disclosed embodiments can be understood and effected by those skilled in the art and practicing the claimed disclosure, from a study of the drawings, the disclosure, and the appended claims.

As used herein, including in the claims, singular forms of terms are to be construed as also including the plural form and vice versa, unless the context indicates otherwise. Thus, it should be noted that as used herein, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise.

The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to fulfill aspects of the present invention. The present technology is also understood to encompass the exact terms, features, numerical values or ranges etc., if in here a relative term, such as "about", "substantially", "ca.", "generally", "at least", "at the most" or "approximately" is used in this specification, such a term should also be construed to also include the exact term. That is, e.g., "substantially straight" should be construed to also include "(exactly) straight". In other words, "about 3" shall also comprise "3" or "substantially perpendicular" shall also comprise "perpendicular". Any reference numerals in the claims should not be considered as limiting the scope.

In the claims, the terms "comprises/comprising", "including", "having", and "contain" and their variations should be understood as meaning "including but not limited to", and are not intended to exclude other components. Furthermore, although individually listed, a plurality of means, elements or method steps may be implemented. Additionally, although individual features may be included in different claims, these may possibly advantageously be combined, and the inclusion in different claims does not imply that a combination of features is not feasible and/or advantageous. In addition, singular references do not exclude a plurality.

Whenever steps were recited in the above or also in the appended claims, it should be noted that the order in which the steps are recited in this text may be the preferred order, but it may not be mandatory to carry out the steps in the recited order. That is, unless otherwise specified or unless clear to the skilled person, the order in which steps are recited may not be mandatory. That is, when the present document states, e.g., that a method comprises steps (A) and (B), this does not necessarily mean that step (A) precedes step (B), but it is also possible that step (A) is performed (at least partly) simultaneously with step (B) or that step (B) precedes step (A). Furthermore, when a step (X) is said to precede another step (Z), this does not imply that there is no step between steps (X) and (Z). That is, step (X) preceding step (Z) encompasses the situation that step (X) is performed directly before step (Z), but also the situation that (X) is performed before one or more steps (Yl), ..., followed by step (Z). Corresponding considerations apply when terms like "after" or "before" are used.

It will be appreciated that variations to the foregoing embodiments of the invention can be made while still falling within the scope of the invention can be made while still falling within scope of the invention. Features disclosed in the specification, unless stated otherwise, can be replaced by alternative features serving the same, equivalent or similar purpose. Thus, unless stated otherwise, each feature disclosed represents one example of a generic series of equivalent or similar features. Use of exemplary language, such as "for instance", "such as", "for example" and the like, is merely intended to better illustrate the invention and does not indicate a limitation on the scope of the invention unless so claimed. Any steps described in the specification may be performed in any order or simultaneously, unless the context clearly indicates otherwise.

All of the features and/or steps disclosed in the specification can be combined in any combination, except for combinations where at least some of the features and/or steps are mutually exclusive. In particular, preferred features of the invention are applicable to all aspects of the invention and may be used in any combination. Reference numbers and letters appearing between parentheses in the claims, identifying features described in the embodiments and illustrated in the accompanying drawings, are provided as an aid to the reader as an exemplification of the matter claimed. The inclusion of such reference numbers and letters is not to be interpreted as placing any limitations on the scope of the claims.

Claims

1. System for increasing traffic safety comprising: a. a monitoring component (3) that is configured to monitor traffic participants (10-13); and b. an analyzing component (1) that is configured to predict a range of trajectories (10a-13a) of the traffic participants.

2. System according to the preceding claim wherein the analyzing component is configured to predict the range of trajectories with a frequency of at least 1 per second, preferably at least 100 per second and more preferably at least 15 per second.

3. System according to any one of the preceding claims wherein the range of trajectories (10a-13a) comprise a value representing the likelihood of one or more trajectories.

4. System according to any one of the preceding claims wherein the analyzing component (1) comprises trained software that has been trained with respect to the likelihood of prediction of the trajectories and/or the precision of the trajectories of the traffic participants.

5. System according to one any of the preceding claims wherein the analyzing component (1) comprises a neural network, preferably a deep neural network.

6. System according to any one of the preceding system embodiments further comprising at least one, preferably two or more cameras to capture traffic images of a traffic location.

7. System according the any one of the preceding system embodiments further comprising a trained software for analyzing images and for detecting accidents.

8. System according to any one of the preceding system embodiments further comprising a warning component (5) that is configured to warn traffic participants (10-13) in case their range of trajectories overlap.

9. System according to any one of the preceding system embodiments further comprising at least one, preferably two or more cameras to capture traffic images of a traffic location

10. System according the any one of the preceding system embodiments further comprising a trained software for analyzing images and for detecting accidents

11. System according to any one of the preceding system embodiments further comprising a warning component (5) that is configured to warn traffic participants (10-13) in case their range of trajectories overlap

12. Method for increasing traffic safety comprising: a. monitoring traffic participants (10-13) by a monitoring component (3); and b predicting a range of trajectories (10a-13a) of the traffic participants by an analyzing component (1).

13. Method according to claim 12 wherein the range of trajectories (10a-13a) comprises a value representing the likelihood of one or more trajectories.

14. Method according to the claims 12 and 13 further with the step of training software of the analyzing component (1) with respect to the likelihood of prediction of the trajectories and/or the precision of the trajectories of the traffic participants.

15. Method according to any of the claims 12 to 14 wherein the analyzing component (1) comprises a neural network or a deep neural network.