Emergency escape method, device and equipment for vehicle and medium

Abstract

The embodiment of the invention provides an emergency escape method, device and equipment for a vehicle and a medium. The method comprises the steps that a dangerous source existing in the vehicle, dangerous information of the dangerous source and the safety score of each initial escape exit of the vehicle are determined according to current vehicle internal environment information; according to the passenger state information and the danger information, the accumulated passing cost of the passenger from the current compartment position to each initial escape exit is determined; and determining a target escape exit according to the accumulated passing cost and the safety score, and sending multi-mode guidance information to passengers. According to the embodiment of the invention, by sensing the environment in the vehicle and the passenger state in real time, the personalized escape exit of the passenger is dynamically planned according to the environment information and the passenger state, the method adapts to complex dangerous cases and individual differences, and the passenger is guided to escape through the multi-mode guide information, so that the escape success rate is effectively improved.

Description

Technical Field

[0001] This application relates to the field of emergency escape technology, and in particular to an emergency escape method, device, equipment, and medium for vehicles. Background Technology

[0002] With the widespread use of automobiles, vehicle fires have become a significant safety risk. Unlike traditional fires, vehicle interiors are enclosed spaces with concentrated flammable materials, limited escape exits, and may be accompanied by a combination of hazards such as toxic gases (e.g., carbon monoxide) and battery explosions, leaving occupants with extremely limited escape time.

[0003] Existing vehicle fire emergency escape methods typically rely on controlling door unlocking based on fixed physical thresholds, such as specific temperatures or gas concentrations. These methods are slow to respond and cannot address complex situations where the fire source location is variable and the occupants' conditions vary. Furthermore, their response patterns are usually uniform, such as unlocking all doors simultaneously, failing to provide differentiated escape guidance for occupants in different locations and potentially leading them into more dangerous areas. Additionally, they typically use constantly lit emergency lights or uniform audible and visual alarms to indicate fixed directions, failing to adjust the guidance direction according to the dynamically spreading danger and failing to consider individual differences among occupants. Summary of the Invention

[0004] This application provides an emergency escape method, device, equipment, and medium for vehicles, aiming to improve the existing emergency escape methods that cannot cope with complex and ever-changing dangerous situations and do not take into account the occupant status.

[0005] To address the aforementioned problems, this application discloses an emergency escape method for vehicles, the method comprising: Obtain current environmental information inside the vehicle and occupant status information; Based on the environmental information, determine the hazards present inside the vehicle, the hazard information of the hazards, at least two initial escape exits of the vehicle, and the safety score of each initial escape exit; Based on the occupant status information and the danger information, determine the cumulative passage cost for the occupant from the current carriage position to each of the initial escape exits; The target escape exit is selected from the initial escape exits based on the cumulative passage cost and the safety score; Multimodal guidance information is sent to the occupants based on the target escape exit.

[0006] In some feasible implementations, the occupant status information includes occupant position, occupant age category, occupant posture, and occupant capability coefficient, and the method includes: Acquire image data of the vehicle's interior; Target recognition and pose estimation are performed on the image data to obtain the occupant's position, age category, and pose. The occupant's ability is assessed using the occupant's age category and posture to obtain the occupant's occupant ability coefficient.

[0007] In some feasible implementations, determining the cumulative travel cost for the occupant from their current position to each of the initial escape exits based on the occupant status information and the hazard information includes: Based on the aforementioned hazard information, the passage cost for each compartment of the vehicle is calculated. Obtain personalized path planning algorithms; The personalized path planning algorithm is used to plan paths for each of the initial escape exits based on the occupant's position, occupant's capacity coefficient, and the passage cost of each carriage, so as to obtain candidate paths for each of the initial escape exits. The cumulative passage cost of the initial escape exit is calculated based on the candidate path and the passage cost.

[0008] In some feasible implementations, the hazard information includes the hazard intensity per unit distance of the hazard source, and the calculation of the passage cost for each compartment position of the vehicle based on the hazard information includes: Calculate the hazardous distance between the location of the carriage and the hazard source; The danger index of the carriage location is calculated based on the danger intensity per unit distance and the danger distance. Determine the location category of the carriage and obtain the basic passage cost corresponding to the location category; The passage cost for the carriage location is calculated based on the basic passage cost and the danger index.

[0009] In some feasible implementations, the method further includes: Obtain the characteristic information of the hazard source, which includes at least the maximum temperature, smoke concentration, toxic gas concentration, open flame indication, and battery failure level; The maximum temperature, smoke concentration, toxic gas concentration, open flame indication, and battery fault level are weighted and fused to obtain the hazard intensity per unit distance of the hazard source.

[0010] In some feasible implementations, the environmental information includes at least temperature, smoke concentration, and toxic gas concentration. The step of determining, based on the environmental information, the presence of hazards inside the vehicle, hazard information of the hazards, at least two initial escape exits of the vehicle, and safety scores for each initial escape exit includes: Based on the temperature, smoke concentration, and toxic gas concentration, a hazard assessment is performed on each compartment of the vehicle to obtain a comprehensive hazard index for each compartment location. If the comprehensive hazard index is greater than the preset hazard threshold, the location of the carriage is marked as a hazard source, and the coordinate information, comprehensive hazard index, and hazard intensity per unit distance of the hazard source are obtained. Obtain the initial escape exit of the vehicle and calculate the first distance between the initial escape exit and the hazard source; The safety score of the initial escape exit is calculated based on the first distance and the hazard intensity per unit distance of the hazard source.

[0011] In some feasible implementations, the step of issuing multimodal guidance information to the occupants based on the target escape exit includes: Obtain the escape path corresponding to the target escape exit; Based on the escape route, multimodal guidance information is issued to the occupants, including at least one of voice prompts, light prompts, and text prompts.

[0012] In some feasible implementations, the method further includes: The cumulative passage cost of the target escape exit, the safety score of the target escape exit, and the occupant status information are detected in real time to obtain real-time detection information; Based on the real-time detection information and preset replanning conditions, it is determined whether to replan the escape route. If so, the escape path is modified to obtain the target escape path.

[0013] This application also discloses an emergency escape device for a vehicle, the device comprising: The information acquisition module is used to acquire current environmental information inside the vehicle and occupant status information; An environmental perception module is used to determine, based on the environmental information, the presence of hazards inside the vehicle, the hazard information of the hazards, at least two initial escape exits of the vehicle, and the safety score of each initial escape exit. The passage cost calculation module is used to determine the cumulative passage cost of the occupant from the current position in the carriage to each of the initial escape exits based on the occupant status information and the danger information; An escape exit determination module is used to select a target escape exit from the initial escape exits based on the cumulative passage cost and the safety score; The escape guidance module is used to send multimodal guidance information to the occupants based on the target escape exit.

[0014] In some feasible implementations, the apparatus further includes: The image acquisition module is used to acquire image data of the interior of the vehicle; The image processing module is used to perform target recognition and pose estimation on the image data to obtain the occupant's occupant position, occupant age category, and occupant pose. The capability assessment module is used to assess the capabilities of the occupants based on their age category and posture, and to obtain the occupant capability coefficient.

[0015] In some feasible implementations, the passage cost calculation module is specifically used for: Based on the aforementioned hazard information, the passage cost for each compartment of the vehicle is calculated. Obtain personalized path planning algorithms; The personalized path planning algorithm is used to plan paths for each of the initial escape exits based on the occupant's position, occupant's capacity coefficient, and the passage cost of each carriage, so as to obtain candidate paths for each of the initial escape exits. The cumulative passage cost of the initial escape exit is calculated based on the candidate path and the passage cost.

[0016] In some feasible implementations, the hazard information includes the hazard intensity per unit distance of the hazard source, and the passage cost calculation module is specifically used for: Calculate the hazardous distance between the location of the carriage and the hazard source; The danger index of the carriage location is calculated based on the danger intensity per unit distance and the danger distance. Determine the location category of the carriage and obtain the basic passage cost corresponding to the location category; The passage cost for the carriage location is calculated based on the basic passage cost and the danger index.

[0017] In some feasible implementations, the apparatus further includes: The characteristic information acquisition module is used to acquire the characteristic information of the hazard source, which includes at least the maximum temperature, smoke concentration, toxic gas concentration, open flame indicator, and battery failure level. The hazard intensity calculation module is used to weight and fuse the highest temperature, smoke concentration, toxic gas concentration, open flame indication, and battery fault level to obtain the hazard intensity per unit distance of the hazard source.

[0018] In some feasible implementations, the environmental information includes at least temperature, smoke concentration, and toxic gas concentration, and the environmental sensing module includes: The hazard index calculation module is used to perform hazard assessment on each compartment location of the vehicle based on the temperature, smoke concentration and toxic gas concentration, and obtain a comprehensive hazard index for each compartment location. The hazard source marking module is used to mark the carriage location as a hazard source if the comprehensive hazard index is greater than a preset hazard threshold, and to obtain the coordinate information, comprehensive hazard index, and hazard intensity per unit distance of the hazard source; The first distance calculation module is used to obtain the initial escape exit of the vehicle and calculate the first distance between the initial escape exit and the hazard source; The safety assessment module is used to calculate the safety score of the initial escape exit based on the first distance and the hazard intensity per unit distance of the hazard source.

[0019] In some feasible implementations, the escape guidance module is specifically used for: Based on the escape path corresponding to the target escape exit, multimodal guidance information is issued to the occupants, including at least one of voice prompts, light prompts, and text prompts.

[0020] In some feasible implementations, the apparatus further includes: The real-time detection module is used to detect the passage cost of the carriage location, the safety score of the target escape exit, and the occupant status information in real time, and obtain real-time detection information. The replanning judgment module is used to determine whether to replan the escape route based on the real-time detection information and preset replanning conditions. The path replanning module is used to modify the escape path if the condition is met, so as to obtain the target escape path.

[0021] This application also discloses an electronic device, including a processor and a memory, wherein the memory is used to store computer programs; and the processor is used to execute the programs stored in the memory to implement any of the methods described in this application.

[0022] This application also discloses a vehicle that includes the electronic equipment described in the above embodiments.

[0023] This application also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the methods described in this application.

[0024] The embodiments of this application have the following advantages: In this embodiment, the system acquires current environmental information inside the vehicle and occupant status information; based on the environmental information, it determines existing hazards inside the vehicle, hazard information of these hazards, at least two initial escape exits, and safety scores for each initial escape exit; based on the occupant status information and hazard information, it determines the cumulative travel cost from the occupant's current position to each initial escape exit; based on the cumulative travel cost and safety score, it selects a target escape exit from the initial escape exits; and based on the target escape exit, it issues multimodal guidance information to the occupant. This embodiment, by real-time sensing of the in-vehicle environment and occupant status, dynamically plans personalized escape exits for occupants based on the environmental information and occupant status, flexibly responding to complex and ever-changing dangerous situations, considering individual differences, and guiding occupants to escape through multimodal guidance information, providing different guidance perception methods and effectively improving the success rate of escape. Attached Figure Description

[0025] Figure 1 This is a flowchart of an emergency escape method for a vehicle provided in one embodiment of this application; Figure 2 This is a schematic diagram of a vehicle emergency escape scenario provided in one embodiment of this application; Figure 3 This is a structural diagram of an emergency escape device for a vehicle provided in one embodiment of this application; Figure 4 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] To make the technical problems, technical solutions, and beneficial effects solved by this application clearer, the following detailed description is provided in conjunction with embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0027] As an example, existing vehicle fire emergency escape methods fall into three categories: Mechanical triggering schemes, which use low-melting-point alloys or shape-memory metals to trigger door lock opening, rely on fixed physical thresholds such as specific temperatures, exhibit delayed response, and cannot cope with complex situations where the fire source location is variable and the occupants' conditions differ; and Electronic triggering schemes, which use temperature and smoke sensors to trigger electronic door lock unlocking or window breaking, heavily rely on the vehicle's main power supply and central controller. In scenarios where short circuits or control system failures may occur in the early stages of a fire, such schemes risk functional paralysis, and their response patterns are usually uniform, such as all doors unlocking simultaneously, failing to provide differentiated escape guidance for occupants in different locations and potentially leading them into more dangerous areas. Static guidance schemes use constantly lit emergency lights or uniform audible and visual alarms to indicate a fixed direction, unable to adjust the guidance direction according to the dynamically spreading danger, and unable to consider individual differences among occupants.

[0028] Based on this, the embodiments of this application acquire the current environmental information inside the vehicle and the occupant status information; determine the hazards present inside the vehicle, the hazard information of the hazards, at least two initial escape exits, and the safety score of each initial escape exit based on the environmental information; determine the cumulative passage cost for the occupant from the current position in the vehicle to each initial escape exit based on the occupant status information and hazard information; select the target escape exit from the initial escape exits based on the cumulative passage cost and safety score; and issue multimodal guidance information to the occupant based on the target escape exit. This achieves real-time perception of the in-vehicle environment and occupant status, and then dynamically plans personalized escape exits for the occupant based on the environmental information and occupant status, flexibly responding to complex and ever-changing dangers, considering individual differences, and guiding the occupant to escape through multimodal guidance information, providing different guidance perception methods and effectively improving the success rate of escape.

[0029] Reference Figure 1 This application provides a flowchart of a vehicle emergency escape method, which includes the following steps: Step 110: Obtain the current environmental information inside the vehicle and the occupant status information; Optionally, this application can be applied to an emergency escape control system without relying on the vehicle's power supply and central controller, enabling normal operation even in the event of a vehicle power supply or main controller failure. Accordingly, this application is illustrated using an emergency escape control system as an example. It should be noted that this application does not limit the specific implementing entity.

[0030] In this embodiment, multi-source sensors are used to collect real-time environmental information and occupant status information inside the vehicle. Environmental information may include temperature, concentrations of toxic gases such as carbon monoxide, and smoke concentration at various locations within the vehicle. Occupant status information includes at least the occupant's location and physiological and behavioral state. This embodiment allows for real-time perception of the vehicle's environment and occupant status, enabling dynamic planning of escape routes based on real-time information, and flexible responses to complex and changing emergencies and differences in occupant status.

[0031] Step 120: Based on the environmental information, determine the hazards present inside the vehicle, the hazard information of the hazards, at least two initial escape exits of the vehicle, and the safety score of each initial escape exit; In this embodiment, one or more hazardous areas (i.e., hazard sources) inside the vehicle are identified based on collected multi-dimensional environmental information. Hazard information for these hazard sources is determined, and all available escape exits (i.e., initial escape exits) are identified. The safety level of each initial escape exit is then assessed to obtain a safety score. Here, a hazard source indicates that the location of the vehicle compartment is a hazardous area; hazard information includes at least the location and degree of hazard of the hazard source; initial escape exits include doors, windows, sunroofs, etc.; and the safety score represents the safety level of escaping from the initial escape exit under the current circumstances. The hazard source, hazard information, available escape exits, and the safety score of the escape exit constitute the environmental situation inside the vehicle. This environmental situation allows for a direct and comprehensive understanding of the vehicle's internal escape capabilities. Therefore, this embodiment generates an environmental situation characterizing the hazardous and escape conditions inside the vehicle based on environmental information, which helps in the subsequent accurate planning of occupant escape routes.

[0032] Step 130: Based on the occupant status information and the danger information, determine the cumulative passage cost for the occupant from the current carriage position to each of the initial escape exits; In this embodiment, based on the occupant's status information and the hazard information of various hazards inside the vehicle, the cumulative travel cost for the occupant to move from their current location to each initial escape exit is calculated. The cumulative travel cost represents the total cost or effort incurred by the occupant during their movement from their current location to the initial escape exit; it can also be understood as the potential danger posed to the occupant by the hazards during their movement. This embodiment selects escape exits by combining the hazard situation at various locations inside the vehicle with the occupant's status information, thus achieving escape path planning that considers the occupant's status and meets the needs of individual differences.

[0033] Step 140: Select a target escape exit from the initial escape exits based on the cumulative passage cost and the safety score; In this embodiment, based on the cumulative passage cost and safety score of each initial escape exit, the target escape exit with the highest probability of successful escape is selected from multiple initial escape exits. This embodiment effectively improves the occupant's escape success rate by combining the escape cost and safety level of each escape exit to determine the exit with the highest success rate.

[0034] Step 150: Send multimodal guidance information to the occupants based on the target escape exit.

[0035] In this embodiment, after determining the target escape exit, multimodal guidance information is sent to the occupants in real time based on the target escape exit. This multimodal guidance information guides the occupants to move to the target escape exit and may include, but is not limited to, voice prompts, light prompts, and text prompts. This embodiment provides multiple perception methods through multimodal guidance information, enabling occupants to intuitively perceive the escape route while also catering to the perception needs of different occupants, effectively improving the escape success rate.

[0036] This application embodiment senses the in-vehicle environment and occupant status in real time, and then dynamically plans personalized escape exits and escape routes for occupants based on environmental information and occupant status. It flexibly responds to complex and ever-changing dangers, takes into account individual differences, and guides occupants to escape through multimodal guidance information, providing different guidance perception methods and effectively improving the success rate of escape.

[0037] In some feasible implementations, the occupant status information includes occupant position, occupant age category, occupant posture, and occupant capability coefficient, and the method includes: Acquire image data of the vehicle's interior; Target recognition and pose estimation are performed on the image data to obtain the occupant's position, age category, and pose. The occupant's ability is assessed using the occupant's age category and posture to obtain the occupant's occupant ability coefficient.

[0038] In this embodiment, the occupant status information includes at least occupant position, occupant age category (e.g., elderly, youth, children), occupant posture (e.g., upright, lying down), and occupant capability coefficient. The occupant capability coefficient characterizes the occupant's escape ability. By identifying the vehicle's interior image data, the position and age category of each occupant are obtained. The occupant's posture is further estimated to obtain posture information. Finally, the occupant's escape ability is assessed based on their age category and posture to obtain the occupant capability coefficient. By acquiring occupant status information, this application helps to dynamically plan escape routes based on occupant status, taking into account individual occupant differences and effectively improving the escape success rate.

[0039] As an example, firstly, raw image frames I are captured by an in-vehicle camera, and then the seat pressure values ​​P of each seat are obtained by the seat pressure sensors. k (k=1, 2...x, where x equals the number of seats).

[0040] Secondly, a pre-trained lightweight convolutional neural network is used to perform real-time object detection on the original image frame I, outputting the bounding box (BBox) for each occupant. m In addition, the system categorizes occupants by age, classifying them as "elderly," "youth," or "children," and uses camera calibration parameters to define bounding boxes (BBoxes). m Transform the center point to a two-dimensional position in the vehicle coordinate system (x m y m The system obtains the occupant's position and uses an attitude estimation model to determine whether the occupant's attitude is "upright" or "tilted".

[0041] Furthermore, the identified occupant location (x) m y m The system matches the seat location activated by the pressure sensor (i.e., the seat where an occupant is identified based on the seat pressure value) with the occupant's position to cross-validate the validity of the occupant status information. For example, if the image recognizes an occupant in the left rear seat and the left rear seat pressure sensor P... k If the pressure threshold is reached, the occupant's status information is confirmed to be valid.

[0042] Furthermore, based on the age category and posture of the occupants, a capability coefficient C is assigned to each occupant m. m C m Determined by the base age coefficient and posture correction coefficient: C m =Base age coefficient × Posture correction coefficient. The base age coefficient includes a base ability coefficient of 0.5-0.7 for children, a base ability coefficient of 1 for young adults, and a base ability coefficient of 0.6-0.8 for the elderly. The base age coefficient can be further subdivided and adjusted according to age. The posture correction coefficient includes a normal upright sitting posture of 1, a tilted or curled-up sitting posture that may have been attacked of 0.7-0.9, and a tilted or rolling state of 0.3-0.5.

[0043] Finally, define a crew list O, which contains each crew member's {ID, position (x, y), ability coefficient C, and status label}. The status label includes at least the crew member's age category and current posture, and may also include the crew member's state of consciousness, visible external injuries, etc.

[0044] In some feasible implementations, determining the cumulative travel cost for the occupant from their current position to each of the initial escape exits based on the occupant status information and the hazard information includes: Based on the aforementioned hazard information, the passage cost for each compartment of the vehicle is calculated. Obtain personalized path planning algorithms; The personalized path planning algorithm is used to plan paths for each of the initial escape exits based on the occupant's position, occupant's capacity coefficient, and the passage cost of each carriage, so as to obtain candidate paths for each of the initial escape exits. The cumulative travel cost of the candidate path is calculated based on the candidate path and the travel cost.

[0045] In this embodiment, firstly, based on the hazard information of the dangerous sources inside the vehicle, an escape assessment is performed on each compartment location inside the vehicle to obtain the passage cost of each compartment location. The passage cost represents the cost or expense that the occupant needs to bear when passing through that compartment location, which can also be understood as the risk that needs to be borne, such as reduced visibility, heat radiation, etc. By calculating the passage cost of each compartment location, a risk field inside the vehicle is constructed, which helps to fully perceive the risks existing in each location inside the vehicle.

[0046] Furthermore, a personalized path planning algorithm is used to plan paths based on occupant positions, occupant capability coefficients, and the passage costs at each carriage position. The path with the minimum passage cost at each initial escape exit is output as a candidate path; that is, the optimal path at each initial escape exit is taken as the candidate path, and the sum of the passage costs at each carriage position along the candidate path is calculated as the cumulative passage cost of that initial escape exit. This embodiment generates personalized escape paths for each escape exit and corresponding cumulative passage costs based on occupant status information and the risk field inside the vehicle.

[0047] As an example, firstly, the impact of each hazard on the location of the carriage is determined based on the hazard information of each hazard, thereby determining the passage cost of the carriage location; Secondly, an improved A* algorithm is used to find the starting point Start for occupant m. m This refers to the path from the passenger's current location in the carriage to each available exit that minimizes both the travel cost and the risk exposure time. F(n) = G(n) + W(m) * H(n); W(m) = λ / C m ; Where G(n) is the value starting from the starting point Start m The actual cumulative travel cost to the current node n (i.e., the sum of travel costs for each carriage position on the path), H(n) is a heuristic function, is the Euclidean distance from the current node n to the initial escape exit, W(m) is a personalized weighting factor, and C m λ is the crew capacity coefficient, and λ is the normalization constant.

[0048] For crew members with weaker abilities (C) mFor younger children or the elderly, as W(m) increases, the algorithm tends to choose a shorter path with a smaller H(n), even if the cost of travel may be slightly higher, to reduce the occupants' exposure time in hazardous environments; for more capable occupants, the algorithm tends to choose a lower-cost, safer path, but which may be slightly longer.

[0049] Furthermore, parallel computing starts from the beginning. m The cumulative passage cost F corresponding to the path to each available exit total(exit) .

[0050] Finally, select the option with the minimum cumulative passage cost min{F}. total(exit_j) The exit of} is used as the target escape route.

[0051] In some feasible implementations, the hazard information includes the hazard intensity per unit distance of the hazard source, and the calculation of the passage cost for each compartment position of the vehicle based on the hazard information includes: Calculate the hazardous distance between the location of the carriage and the hazard source; The danger index of the carriage location is calculated based on the danger intensity per unit distance and the danger distance. Determine the location category of the carriage and obtain the basic passage cost corresponding to the location category; The passage cost for the carriage location is calculated based on the basic passage cost and the danger index.

[0052] In this embodiment, the hazard information of the hazard source also includes the hazard intensity per unit distance for each hazard source. Specifically, the hazard distance between the carriage location and the hazard source is calculated, and then the hazard index of the hazard source for the carriage location is calculated based on the hazard distance and the hazard intensity per unit distance of the hazard source. If there are multiple hazard sources, the hazard indices of multiple hazard sources for the carriage location need to be superimposed to determine the final hazard index of the carriage location, or the highest hazard index is selected as the final hazard index of the carriage location. Further, the location category of the carriage location is determined, such as whether the carriage location is flat ground or whether there are obstacles such as seats in the carriage location. Different categories of carriage locations are set with different basic passage costs. Finally, the passage cost of the carriage location is calculated based on the basic passage cost and the hazard index of the carriage location.

[0053] As an example, firstly, the overhead view of the vehicle is discretized into an M*N grid (e.g., 10cm*10cm), where each cell (Cell(i,j)) contains its coordinates and attributes; further, the passage cost for each cell is calculated to generate a cost map of the vehicle. Cost(i,j)=BaseCost+K*H(i,j); Where BaseCost is the basic passage cost (1 for flat ground, and infinite for obstacles), H(i,j) is the danger index of the carriage position, and K is the amplification factor.

[0054] As an example, the hazard index H is calculated based on either a Gaussian decay model or an exponential decay model. Taking the Gaussian model as an example: For a single hazard source k, the hazard value hk(x,y) it causes at any point (x,y) in space (i.e., the position of the carriage) is: h k(x,y) =w k *exp(-(d 2 ) / (2*σ 2 )); d=sqrt((xx k ) 2 +(yy k ) 2 ); Where (x, y) represents the distance from the carriage to the hazard source k, and σ is the attenuation coefficient, controlling the range of influence. The larger σ is, the wider the hazard spread. That is, at the center of the hazard source (d=0), the hazard value is equal to w. k The danger value decreases smoothly as the distance d increases.

[0055] The total hazard index of a point (x, y) in space is the maximum value or sum of the effects of all hazard sources at that point. That is, for each carriage position (i, j), the coordinates (x, y) of its center point are taken. i ,y j ), calculate H max(xi,yj) Or H add(xi,yj) The danger index H(i,j) of the carriage location is obtained.

[0056] In some feasible implementations, the method further includes: Obtain the characteristic information of the hazard source, which includes at least the maximum temperature, smoke concentration, toxic gas concentration, open flame indication, and battery failure level; The maximum temperature, smoke concentration, toxic gas concentration, open flame indication, and battery fault level are weighted and fused to obtain the hazard intensity per unit distance of the hazard source.

[0057] In this embodiment, the characteristic information of the hazard source includes at least the maximum temperature (the highest temperature at the location of the hazard source), smoke concentration (interpolated or averaged values ​​of smoke sensors at the location of the hazard source), toxic gas concentration (interpolated or averaged values ​​of gas sensors at the location of the hazard source), open flame indicator (e.g., 1 indicates visible open flame, 0 indicates no visible open flame), and battery fault level (e.g., 0 for normal, 1 for warning, 2 for thermal runaway). Based on this characteristic information, the hazard intensity per unit distance of the hazard source is determined. This embodiment comprehensively and accurately determines the degree of hazard impact of the hazard source through multi-dimensional information.

[0058] As an example, firstly, for each characteristic index—maximum temperature T, smoke concentration S, toxic gas concentration C, open flame indicator F, and battery fault level B—different functions f are designed based on their own characteristics. T f S f C f F f B This is used to map raw data of different characteristic indicators to a unified interval of [0, 1], where 0 represents no threat and 1 represents reaching a preset critical threat. For example, temperature is mapped using an S-shaped function to reflect the critical point effect of temperature, while smoke and gas are mapped using a linear saturation function.

[0059] Furthermore, weights are set for different characteristic indicators, such as open flame weight d = 0.35, temperature weight a = 0.25, gas concentration weight c = 0.2, smoke concentration weight b = 0.15, and battery fault level weight e = 0.05. All weights can be added together to equal 1. This application does not impose specific restrictions on this.

[0060] Finally, the various characteristic indicators are weighted and fused to obtain the hazard intensity per unit distance of the hazard source: w k = a * f T + b * f S +c * f C + d * f F + e* f B ; Optionally, since the danger occurs dynamically, w can be adjusted. k Dynamic time correction can be performed, for example: when the rate of change Δw > 0, a gain coefficient η can be set, such as w k_now =w k *(1+η*Δw), amplifies the trend of escalating danger.

[0061] In some feasible implementations, the environmental information includes at least temperature, smoke concentration, and toxic gas concentration. The step of determining, based on the environmental information, the presence of hazards inside the vehicle, hazard information of the hazards, at least two initial escape exits of the vehicle, and safety scores for each initial escape exit includes: Based on the temperature, smoke concentration, and toxic gas concentration, a hazard assessment is performed on each compartment of the vehicle to obtain a comprehensive hazard index for each compartment location. If the comprehensive hazard index is greater than the preset hazard threshold, the location of the carriage is marked as a hazard source, and the coordinate information, comprehensive hazard index, and hazard intensity per unit distance of the hazard source are obtained. Obtain the initial escape exit of the vehicle and calculate the first distance between the initial escape exit and the hazard source; The safety score of the initial escape exit is calculated based on the first distance and the hazard intensity per unit distance of the hazard source.

[0062] In this embodiment, the specific process for generating the environmental situation is as follows: Multi-dimensional data from each carriage location is integrated to assess the degree of danger at that location, obtaining a comprehensive danger index. If the comprehensive danger index of a carriage location is greater than a danger threshold, the carriage location is marked as a danger source. The coordinate information and comprehensive danger index of the carriage location are then obtained to generate danger information for that danger source. Furthermore, all available exits of the vehicle are obtained as initial escape exits, and the safety score of the initial escape exit is calculated based on the distance between the unit distance danger intensity of each danger source and the distance between the initial escape exit. This embodiment generates a global environmental situation characterizing the danger and escape conditions inside the vehicle based on environmental information, which helps in subsequent escape route planning.

[0063] As an example, the system receives the three-dimensional coordinates (x, y) of multiple temperature sensors, smoke sensors, and gas sensors located inside the vehicle. i y i , z i The data includes temperature (T), smoke concentration (S), and toxic gas concentration (C), and uses Kriging interpolation or linear interpolation to construct a continuous three-dimensional environmental field M based on discrete sensor points. heat(x，y，z) This creates a heat map.

[0064] Secondly, calculate the overall hazard index H for each location: H(x, y, z) = α*T norm(x，y，z) +β*S norm(x，y，z) +γ*C norm(x，y，z) ; Among them, T norm S norm Cnorm These are the normalized values ​​for temperature, smoke concentration, and toxic gas concentration. α, β, and γ are weighting coefficients, which can be calibrated according to requirements, such as α=0.5, β=0.3, and γ=0.2.

[0065] Furthermore, areas with H values ​​greater than a threshold (such as 0.7) are marked as hazard sources.

[0066] Furthermore, the safety score (Sexit_j) for each initial escape exit is calculated based on the sum of the weighted inverses of the distances from the initial escape exit to all hazards: S exit_j_init =Σ(w k / (dist(j,Source k )+ε)); Where, (dist(j, Source) k Let ) be the distance from exit j to hazard source k, and w be the distance from exit j to hazard source k. k Let ε be the hazard intensity per unit distance from hazard source k, where ε is a very small positive number to prevent the denominator from being zero (when the exit is right on the hazard source), such as 0.0001. The farther the distance, the higher the fraction, which means the safer it is.

[0067] Finally, the environmental situation E is generated, which includes a list of hazard sources k (the center coordinates and comprehensive hazard index of each hazard source) and an initial safety score S for each escape exit j. exitj .

[0068] In some feasible implementations, the step of issuing multimodal guidance information to the occupants based on the target escape exit includes: Based on the escape path corresponding to the target escape exit, multimodal guidance information is issued to the occupants, including at least one of voice prompts, light prompts, and text prompts.

[0069] In this embodiment of the application, after determining the target escape exit for occupants, multimodal guidance information will be issued to the occupants based on the escape path corresponding to the target escape exit (i.e. the path with the minimum cumulative passage cost required for the occupants to move from their current location to the target escape exit), guiding the occupants to move to the target escape exit according to the escape path.

[0070] For example, based on the occupant's real-time location, targeted and progressive voice commands may be issued, and / or, an LED light array or light strip arranged inside the vehicle may be controlled to form a bright path extending from the occupant's current location to the target escape exit, and / or, text prompts may be displayed on a screen inside the vehicle.

[0071] In some feasible implementations, the method further includes: The passage cost of the carriage location, the safety score of the target escape exit, and the occupant status information are detected in real time to obtain real-time detection information; Based on the real-time detection information and preset replanning conditions, it is determined whether to replan the escape route. If so, the escape path is modified to obtain the target escape path.

[0072] In this embodiment of the application, the changes in the passage cost of each carriage position, the changes in the safety score of the target escape exit, and the changes in the occupant status information will also be detected in real time. If the replanning conditions are met, the escape route will be replanned. The replanning conditions include at least the growth rate or absolute value of the passage cost of the carriage position on the escape route, the safety score of the target escape exit, and the movement status of the occupants.

[0073] Specifically, if the growth rate or absolute value of the passage cost at any position in the escape route exceeds a preset threshold, and / or, the safety score of the target escape exit is lower than a preset lower limit, and / or, the occupant stops moving or deviates from the escape route, and / or, the occupant's condition deteriorates (such as the occupant's ability coefficient dropping sharply due to changes in posture), then the escape route will be replanned.

[0074] As an example, calculate the cost growth rate ΔCost / Δt for any position in the carriage within a sliding time window (e.g., the most recent second). If ΔCost / Δt > a threshold, it is determined that the fire or smoke is spreading rapidly at that location, the escape route is about to become invalid, and the escape route needs to be replanned.

[0075] If the cost of passing through any carriage exceeds the absolute threshold, it indicates that the carriage is no longer passable and an escape route needs to be replanned.

[0076] If the safety score of the target escape exit is lower than the lower threshold, such as dropping from 0.8 to 0.3, it means that the exit is no longer safe due to the spread of fire, external obstacles, or failure to break windows, and the escape route needs to be replanned.

[0077] Monitoring the real-time location Pos of occupant m m(t) and speed V m(t) If V m(t) If the duration T of the speed threshold (e.g., 0.1 m / s) exceeds the first time threshold (e.g., 5 seconds), the occupant is determined to be stationary and may be injured, panicked, or trapped, requiring a replanning of the escape route.

[0078] Calculate the vertical distance D from the occupant's current position to the center line of the planned path. If D > distance threshold (e.g., 1 meter) and duration T > second time threshold (e.g., 3 seconds), it is determined that the occupant has seriously deviated from the escape path, possibly due to unclear guidance, avoidance of personal danger, etc., and the escape path needs to be replanned.

[0079] If the occupant's C m If a person falls precipitously due to a change in posture (such as from "sitting" to "tilting"), it is determined that they have lost their ability to act independently, and it is necessary to replan their escape route.

[0080] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, the following examples are provided for illustrative purposes: refer to Figure 2 This application provides a vehicle emergency escape method, which specifically includes the following steps: S1. Collect environmental information such as current temperature, smoke concentration, and toxic gas concentration from multi-source sensors inside the vehicle, and use interpolation to generate an environmental heat map inside the vehicle to form a disaster perception network. Further, determine the presence of hazardous sources in the vehicle and the safety scores of each available exit through environmental information to generate the environmental situation of the vehicle. At the same time, perceive the age category and posture of the occupants based on multi-source sensors and calculate the occupant's ability coefficient. S2. Based on the environmental situation and occupant capacity coefficient, generate the path with the minimum passage cost for each available exit as a candidate path. S3. Determine the target escape exit based on the cumulative passage cost of each candidate path and the safety score of the available exits; S4. Unlock or blow up the target escape exit, and send out multimodal guidance information for the escape path corresponding to the target escape exit, while initiating a communication distress call; S5. Real-time detection and determination of whether to replan the escape route based on replanning conditions.

[0081] This application also provides an emergency escape device 30 for a vehicle, please refer to... Figure 3 ,include: The information acquisition module 310 is used to acquire the current environmental information inside the vehicle and the occupant status information; The environmental perception module 320 is used to determine, based on the environmental information, the presence of hazards inside the vehicle, the hazard information of the hazards, at least two initial escape exits of the vehicle, and the safety score of each initial escape exit. The passage cost calculation module 330 is used to determine the cumulative passage cost of the occupant from the current position in the carriage to each of the initial escape exits based on the occupant status information and the danger information. Escape exit determination module 340 is used to select a target escape exit from the initial escape exits based on the cumulative passage cost and the safety score; The escape guidance module 350 is used to send multimodal guidance information to the occupants based on the target escape exit.

[0082] In some feasible implementations, the apparatus further includes: The image acquisition module is used to acquire image data of the interior of the vehicle; The image processing module is used to perform target recognition and pose estimation on the image data to obtain the occupant's occupant position, occupant age category, and occupant pose. The capability assessment module is used to assess the capabilities of the occupants based on their age category and posture, and to obtain the occupant capability coefficient.

[0083] In some feasible implementations, the passage cost calculation module 330 is specifically used for: Based on the aforementioned hazard information, the passage cost for each compartment of the vehicle is calculated. Obtain personalized path planning algorithms; The personalized path planning algorithm is used to plan paths for each of the initial escape exits based on the occupant's position, occupant's capacity coefficient, and the passage cost of each carriage, so as to obtain candidate paths for each of the initial escape exits. The cumulative passage cost of the initial escape exit is calculated based on the candidate path and the passage cost.

[0084] In some feasible implementations, the hazard information includes the hazard intensity per unit distance of the hazard source, and the passage cost calculation module 330 is specifically used for: Calculate the hazardous distance between the location of the carriage and the hazard source; The danger index of the carriage location is calculated based on the danger intensity per unit distance and the danger distance. Determine the location category of the carriage and obtain the basic passage cost corresponding to the location category; The passage cost for the carriage location is calculated based on the basic passage cost and the danger index.

[0085] In some feasible implementations, the apparatus further includes: The characteristic information acquisition module is used to acquire the characteristic information of the hazard source, which includes at least the maximum temperature, smoke concentration, toxic gas concentration, open flame indicator, and battery failure level. The hazard intensity calculation module is used to weight and fuse the highest temperature, smoke concentration, toxic gas concentration, open flame indication, and battery fault level to obtain the hazard intensity per unit distance of the hazard source.

[0086] In some feasible implementations, the environmental information includes at least temperature, smoke concentration, and toxic gas concentration, and the environmental sensing module 320 includes: The hazard index calculation module is used to perform hazard assessment on each compartment location of the vehicle based on the temperature, smoke concentration and toxic gas concentration, and obtain a comprehensive hazard index for each compartment location. The hazard source marking module is used to mark the carriage location as a hazard source if the comprehensive hazard index is greater than a preset hazard threshold, and to obtain the coordinate information, comprehensive hazard index, and hazard intensity per unit distance of the hazard source; The first distance calculation module is used to obtain the initial escape exit of the vehicle and calculate the first distance between the initial escape exit and the hazard source; The safety assessment module is used to calculate the safety score of the initial escape exit based on the first distance and the hazard intensity per unit distance of the hazard source.

[0087] In some feasible implementations, the escape guidance module is specifically used for: Based on the escape path corresponding to the target escape exit, multimodal guidance information is issued to the occupants, including at least one of voice prompts, light prompts, and text prompts.

[0088] In some feasible implementations, the apparatus further includes: The real-time detection module is used to detect the passage cost of the carriage location, the safety score of the target escape exit, and the occupant status information in real time, and obtain real-time detection information. The replanning judgment module is used to determine whether to replan the escape route based on the real-time detection information and preset replanning conditions. The path replanning module is used to modify the escape path if the condition is met, so as to obtain the target escape path.

[0089] This application also provides an electronic device 40, please refer to... Figure 4 It includes a processor 410 and a memory 420, wherein the memory 410 is used to store computer programs; and the processor 420 is used to execute the programs stored in the memory 410 to implement the method described in any of the embodiments of this application.

[0090] This application also discloses a vehicle that includes the electronic equipment described in the above embodiments.

[0091] This application also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the methods described in this application.

[0092] In this application, "multiple" refers to two or more.

[0093] In this application, unless otherwise expressly defined, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0094] The terms “first,” “second,” “third,” “fourth,” etc., in this application (if present) are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0095] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0096] Unless otherwise specified, all steps in this application may be performed sequentially or randomly. For example, if the method includes steps A and B, it means that the method may include steps A and B performed sequentially, or it may include steps B and A performed sequentially. For example, if the method may also include step C, it means that step C may be added to the method in any order. For example, the method may include steps A, B, and C, or it may include steps A, C, and B, or it may include steps C, A, and B, etc.

[0097] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An emergency escape method for a vehicle, characterized in that, include: Obtain current environmental information inside the vehicle and occupant status information; Based on the environmental information, determine the hazards present inside the vehicle, the hazard information of the hazards, at least two initial escape exits of the vehicle, and the safety score of each initial escape exit; Based on the occupant status information and the danger information, determine the cumulative passage cost for the occupant from the current carriage position to each of the initial escape exits; The target escape exit is selected from the initial escape exits based on the cumulative passage cost and the safety score; Multimodal guidance information is sent to the occupants based on the target escape exit.

2. The method according to claim 1, characterized in that, The occupant status information includes occupant position, occupant age category, occupant posture, and occupant capability coefficient; the method includes: Acquire image data of the vehicle's interior; Target recognition and pose estimation are performed on the image data to obtain the occupant's position, age category, and pose. The occupant's ability is assessed using the occupant's age category and posture to obtain the occupant's occupant ability coefficient.

3. The method according to claim 2, characterized in that, The step of determining the cumulative passage cost for each occupant from their current position to each of the initial escape exits based on the occupant status information and the danger information includes: Based on the aforementioned hazard information, the passage cost for each compartment of the vehicle is calculated. Obtain personalized path planning algorithms; The personalized path planning algorithm is used to plan paths for each of the initial escape exits based on the occupant's position, occupant's capacity coefficient, and the passage cost of each carriage, so as to obtain candidate paths for each of the initial escape exits. The cumulative passage cost of the initial escape exit is calculated based on the candidate path and the passage cost.

4. The method according to claim 3, characterized in that, The hazard information includes the hazard intensity per unit distance of the hazard source, and the calculation of the passage cost for each compartment of the vehicle based on the hazard information includes: Calculate the hazardous distance between the location of the carriage and the hazard source; The danger index of the carriage location is calculated based on the danger intensity per unit distance and the danger distance. Determine the location category of the carriage and obtain the basic passage cost corresponding to the location category; The passage cost for the carriage location is calculated based on the basic passage cost and the danger index.

5. The method according to claim 4, characterized in that, The method further includes: Obtain the characteristic information of the hazard source, which includes at least the maximum temperature, smoke concentration, toxic gas concentration, open flame indication, and battery failure level; The maximum temperature, smoke concentration, toxic gas concentration, open flame indication, and battery fault level are weighted and fused to obtain the hazard intensity per unit distance of the hazard source.

6. The method according to claim 5, characterized in that, The environmental information includes at least temperature, smoke concentration, and toxic gas concentration. The step of determining, based on the environmental information, the presence of hazards inside the vehicle, hazard information of the hazards, at least two initial escape exits of the vehicle, and safety scores for each initial escape exit includes: Based on the temperature, smoke concentration, and toxic gas concentration, a hazard assessment is performed on each compartment of the vehicle to obtain a comprehensive hazard index for each compartment location. If the comprehensive hazard index is greater than the preset hazard threshold, the location of the carriage is marked as a hazard source, and the coordinate information, comprehensive hazard index, and hazard intensity per unit distance of the hazard source are obtained. Obtain the initial escape exit of the vehicle and calculate the first distance between the initial escape exit and the hazard source; The safety score of the initial escape exit is calculated based on the first distance and the hazard intensity per unit distance of the hazard source.

7. The method according to claim 3, characterized in that, The step of issuing multimodal guidance information to the occupants based on the target escape exit includes: Based on the escape path corresponding to the target escape exit, multimodal guidance information is issued to the occupants, including at least one of voice prompts, light prompts, and text prompts.

8. The method according to claim 7, characterized in that, The method further includes: The passage cost of the carriage location, the safety score of the target escape exit, and the occupant status information are detected in real time to obtain real-time detection information; Based on the real-time detection information and preset replanning conditions, it is determined whether to replan the escape route. If so, the escape path is modified to obtain the target escape path.

9. An electronic device, characterized in that, Including processor and memory, among which Memory, used to store computer programs; A processor for executing a program stored in memory to implement the method described in any one of claims 1-8.

10. A vehicle, characterized in that, It includes the electronic device as described in claim 9.

Images