Method for operating a control device, control device and vehicle
By adopting serial computing and data management methods in the vehicle control device, the shortcomings of the control device in terms of computing performance and storage space are solved, and significant resource savings and cost reduction are achieved.
Patent Information
- Application Number
- CN202380070683.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-10-10
- Filing Date
- 2023-10-05
- Publication Date
- 2025-05-13
AI Technical Summary
Modern vehicle control devices have shortcomings in computing performance and storage space, resulting in waste of resources and increased costs.
By designing a method for controlling devices, the method serializes the computing process and saves data only when necessary, thereby reducing the use of working memory. The method includes providing the measured value and decision value to the processor, calculating part of the results through the processor, and deleting unnecessary data after the calculation is completed.
This method significantly reduces the storage space and computing power of the control device, and can save up to 99% of the storage space and computing power, reducing the cost of the control device.
Smart Images
Figure CN119998686A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for operating a control device and to a control device. The invention also relates to a vehicle. Background Art
[0002] Control devices are ubiquitous in modern vehicles and are used to control and / or regulate nearly all functions of the vehicle.
[0003] Despite the use of fast microprocessors, computing power in control devices is often not available in sufficient quantities or is expensive. Summary of the invention
[0004] It is therefore an object of the present invention to simplify the operation of a control device and thus save computing power and working memory.
[0005] This object is achieved with a method according to claim 1. Furthermore, this object is achieved with a control device according to claim 10 and with a vehicle having such a control device.
[0006] Advantageous embodiments and refinements are the subject matter of the dependent claims.
[0007] The method is used for operating a control device, in particular a control device for determining living beings or objects in an area. The control device has a processor and a working memory, a first interface for receiving measured values and a second interface for outputting results, wherein the method, in particular the analysis of the measured values by the control device, comprises the following steps:
[0008] - providing the corresponding measured values and the decision values to a working memory, wherein partial results are calculated as a function of the corresponding decision values and the measured values by means of a processor and are provided to the working memory, wherein the working memory is respectively designed to receive the decision values, the corresponding partial results and the corresponding measured values;
[0009] The following steps are implemented for i=1 to i=N respectively:
[0010] - calculating further partial results from further measured values, further decision values and the partial results in the respective preceding step;
[0011] Therein, after i=N steps, the corresponding partial results are provided to the second interface for outputting the results.
[0012] In the method described here, i denotes a control variable which runs from 1 to N and is increased by one in each step.
[0013] The measured values are advantageously provided by a sensor. The respective i-th partial result advantageously approximates the respective N-th partial result. Furthermore, a probability can be calculated from the N-th partial result, wherein the probability indicates, for example, whether a living being is located in the area.
[0014] The processor is advantageously designed as a microprocessor, which is advantageously connected to a main memory via a bus system. Alternatively, the main memory can also be integrated in the processor.
[0015] The corresponding interfaces can also be designed as a common interface, which is suitable both for outputting results and for receiving measured values. Advantageously, a plurality of measured values and a plurality of decision values can also be stored in the working memory.
[0016] The invention takes into account that in current analysis routines, a large number of computing operations are performed and that the data generated or required in the process can be loaded into a significant portion of the working memory. In order to be able to keep the working memory small, the invention proposes that the corresponding calculations be designed to be executable serially and that unnecessary values no longer have to be "retained" (ie, stored in the working memory).
[0017] Advantageously, the decision value (β_i) is a number or a vector which can be provided by means of an algorithm with learning capabilities. Advantageously, the decision value is used to convert the corresponding measured value (x_i) into a contribution to the result in the form of a partial result, so that the product of the corresponding measured value and the corresponding decision value respectively produces a partial result. The partial result thus provides a contribution to the result.
[0018] Advantageously, the result is a number which indicates whether the corresponding measured value corresponds to the actual situation. In other words, the result can be a probability as to whether a living being can be detected in the area.
[0019] The result (as the Nth partial result) is advantageously the sum of the partial results p_1 to p_N-1. The result can be converted into a probability by means of a (logistic) function or another S-function, for example.
[0020] The result can indicate whether a fact exists based on the measured values. For example, the measured values of a sensor can detect whether an object or a person is located in an area.
[0021] An advantageous application of the invention is to detect whether a person, in particular a baby, is located in the passenger compartment of a vehicle, preferably in the passenger compartment of a car. Advantageously, it can be determined what type of creature is in the area. Advantageously, the result can be analyzed in such a way that it is determined whether a baby, an adult or a pet has been detected by means of the sensor.
[0022] The measured value is advantageously provided by a sensor and is advantageously a measured value at a certain time. Advantageously, the measured value can also be an n-tuple of individual measured values. Advantageously, the measured value can also correspond to a complex variable, such as a sum over the time variation of a signal.
[0023] By means of the invention described here, it is advantageously possible for the main memory to be less fully utilized and to be designed to be significantly smaller. This saves considerable effort and costs for the control device.
[0024] In an advantageous embodiment of the invention, the calculation of the respective additional partial results is achieved by multiplication of the measured value with the respective decision value, wherein the product is added to the value of the respective partial result in the preceding step. Advantageously, the respective decision value is assigned to the respective measured value.
[0025] The use of decision values is particularly advantageous for the simple calculation of the corresponding partial results. The simple calculation significantly reduces the computational complexity and thus the load on the processor compared to calculations using neural networks.
[0026] In a further advantageous embodiment of the invention, the decision values (β_i; i=1, . . . , N) are determined with the aid of a learning algorithm.
[0027] The decision values are advantageously values in the decision matrix. Alternatively, the decision values β_i can be parameters of a function, in particular a linear function:
[0028]
[0029] A decision matrix is advantageously used to calculate the result from the measured values x_i.
[0030] By providing only one pair of numbers / vectors to the main memory and / or the processor, the control device can be designed with significantly less power or can be operated in a significantly more energy-efficient manner.
[0031] The invention thus makes it possible to reduce the storage space and computing power in the control device. The method described here makes it possible to save up to 99% of the storage space and a significant portion of the computing power compared to conventional methods according to the prior art.
[0032] In a further advantageous embodiment of the invention, the method serves to evaluate measured values of at least one sensor, in particular an ultra-wideband sensor.
[0033] The sensor is advantageously designed as an ultra-wideband receiver and / or an ultra-wideband transmitter. Advantageously, the control device is used to evaluate the output of the ultra-wideband sensor or the plurality of ultra-wideband sensors. Advantageously, the method is used to detect persons in an area.
[0034] By evaluating the measured values of one (or more) ultra-wideband sensors, an otherwise very complex analysis can advantageously be realized in a resource-saving manner.
[0035] In another advantageous embodiment of the invention, after the calculation of the further partial results, the corresponding measured values and the corresponding decision values are deleted from the working memory or from the cache of the processor (in particular from the preceding (calculation) step). Optionally, after the calculation of the further partial results, the corresponding no longer required partial results can also be deleted from the working memory.
[0036] By deleting no longer needed values from the main memory, the area of the main memory required for the method described here can be significantly reduced. Advantageously, the method proposed here can reduce the required storage area of the main memory to one percent.
[0037] In another advantageous embodiment of the invention, the control device has an additional memory, wherein in the corresponding steps, the corresponding decision values are provided from the additional memory to the working memory. The provision of the decision values is either directly from the additional memory to the working memory by means of a processor, which advantageously provides the required measured values and / or decision values for the corresponding steps to the working memory.
[0038] The further memory is advantageously designed as a non-volatile memory, such as a hard disk or a ROM component. The further memory is advantageously used to store decision values and optionally as a buffer memory for measured values. Advantageously, the further memory is designed as a ROM memory. The further memory can either be integrated in the control unit or connected to the control unit via an interface.
[0039] In a further advantageous embodiment of the invention, the corresponding decision values β_i (i=1, . . . N) are stored in the form of a decision matrix or a so-called decision vector in a further memory.
[0040] The decision matrix is advantageously provided by an algorithm with learning capabilities. Advantageously, a particularly simple form of the decision matrix can be formed as a decision matrix β=(β1, ... β N ), In this regard, the decision matrix can be diagonalized by way of example. Advantageously, the method can be used to perform an analysis of the measured values based on the corresponding decision values, without the resources previously required for this being necessary.
[0041] In a further advantageous embodiment of the invention, the calculation is carried out in such a way that a plurality of measured values and a plurality of decision values are provided to the main memory in one step and (further) partial results are provided accordingly for the respective plurality of measured values and decision values.
[0042] Advantageously, such a partial result can be calculated by summing the respective products of the respective measured value x_i and the respective decision value β_i.
[0043] In other words, the calculation of the partial result is achieved as follows:
[0044]
[0045] By calculating such partial results, the computing time is significantly reduced, since the transfer of the corresponding partial results to the memory does not have to be carried out individually for each i=1 to N (or k=1, ..., i). Advantageously, it can be generated for k values 3 to 10, in particular 4 to 7, preferably i=5. In addition, (depending on the design of the processor) the individual steps can be executed in parallel by means of a processor having multiple cores.
[0046] In a further advantageous embodiment of the invention, the presence or position of a living being in an area, in particular in a passenger compartment of a vehicle, is determined by means of the control device.
[0047] Advantageously, the method or the control device is used to determine whether a baby or a pet, such as a dog, has been "forgotten" in a car. A living being can therefore be understood as a baby or a pet. Advantageously, the area can also be the rear seat of the car.
[0048] Such a so-called “child presence detection” can be implemented particularly easily and resource-savingly using the present invention.
[0049] Description of the control device
[0050] The control device has a processor and a working memory and optionally a further memory, and furthermore has a first interface for receiving measured values and a second interface for outputting results, wherein the control device is set up and designed to carry out the method according to the above description.
[0051] Advantageously, the control device is of particularly simple design and can be provided cost-effectively, in particular due to the low memory requirement.
[0052] The use of such a control device in a vehicle is advantageously achieved. In particular, in the vehicle's own control device, the working memory is allocated in a limited manner for cost reasons and for operational safety. Therefore, a control unit for a vehicle is particularly well suited for the above-mentioned application purposes.
[0053] Advantageously, a vehicle having such a control unit is designed as a motor vehicle or a rail vehicle. Advantageously, a vehicle having the control device is used for transporting people. BRIEF DESCRIPTION OF THE DRAWINGS
[0054] The present invention is further described and illustrated below with reference to the accompanying drawings. In the drawings:
[0055] Figure 1 shows an exemplary control device; and
[0056] Figure 2 A possible method diagram is shown. DETAILED DESCRIPTION
[0057] Figure 1 An exemplary control device SE is shown: the control device SE comprises a processor 1 , a main memory 3 and a further memory 5 . The control device SE also comprises an interface 7 for receiving measured values x_i and an interface 9 for outputting results E.
[0058] The further memory 5 comprises an area in which the decision values β_1 to β_N are contained. The decision values β_i are advantageously stored in the form of a decision matrix in the further memory 5. The further memory 5 is advantageously designed as a non-volatile memory, for example a ROM component.
[0059] The decision values β_i are advantageously provided by means of a learning algorithm and advantageously provided to the further memory 5 or the working memory 3 by means of interfaces 7 , 9 .
[0060] The control device SE is designed and arranged to calculate the partial results p_i from the corresponding measured values x_i with the aid of the decision values β_i. The result E can be calculated by adding the partial results p_i. For this purpose, the processor 1 and the working memory 3 are used in particular. The control device SE carries out the following steps for this purpose:
[0061] In a first step, a first measurement value x_1 is provided to the processor 1 .
[0062] The provision of the first measured value x_1 is effected either directly by the interface 7 or by the working memory 3. In addition, a first decision value β_1 is provided to the processor 1. The corresponding decision value β_i is preferably stored in the further memory 5 and is advantageously transferred from the further memory 5 to the working memory 3 and / or the processor 1;
[0063] In the second part of the first step i (steps are numbered 1 here), the processor 1 multiplies the first decision value β_1 by the first measured value x_1. The result of the multiplication is a partial result p_i, wherein the respective partial result p_i is provided to the working memory 3 and stored there.
[0064] In the respective subsequent step (i=2, . . . , N), the partial result p_i in the preceding step is added to the product of the respective further decision value β_i+1 and the respective further measured value x_i+1. The further partial result p_i+1 is accordingly ascertained.
[0065] This step is carried out for i=2 to N, the number N being determined as a function of a plurality of prepared measured values x_i or correspondingly prepared decision values β_i.
[0066] In a final step, the last calculated partial result p_N is advantageously, for example, converted into a probability by introduction into a logic function and / or the current partial result p_N is made available as result E to the second interface 9 .
[0067] Advantageously, after the calculation of the corresponding (further) partial result p_i+1, the values p_1, x_i and β_i of the corresponding step i which are in the working memory due to the previous calculation are deleted. Thus, in the corresponding calculation, only three values, the corresponding measured value x_i, the corresponding decision value β_i and the corresponding partial result p_i (and optionally the further partial result p_i+1), are stored in the working memory 3. In this way, the working memory 3 can be selected to be particularly small and inexpensive.
[0068] Figure 2 A possible method diagram is shown. A calculation is shown, in which a partial result p_i is assigned to each measured value x_i, wherein for all i=1, ...N the following applies: p_i+1=p_i+(β_i*x_i). The final partial result p_N (for i=N) corresponds to the result p_N=E. Advantageously, the final result can be converted into a probability by means of a logic function.
[0069] In an advantageous application of the invention, the corresponding measured values x_i are measured values of a sensor, in particular an ultra-wideband sensor. Advantageously, the measured values x_i and the corresponding development values β_i are transferred to the working memory 3 or the processor 1 for each step i.
[0070] The specific embodiment shown here shows a particularly simple method for operating a control device SE, in particular for detecting an infant in the passenger compartment of a vehicle.
[0071] In summary, the present invention relates to a method for operating a control device SE, a control device SE configured for this purpose, and a vehicle having such a control device SE. The method is used to analyze a measured value x_i, wherein the measured value x_i can be a measurement curve of an ultra-wideband sensor. The analysis is advantageously used to find out whether a creature is located in an area, such as a passenger compartment. The method comprises i steps, which can advantageously be performed successively and in the corresponding step i, the corresponding measured value x_i is combined with a decision value β_i, in particular multiplied, and added to the partial result p_i in the previous step i. The above-mentioned step i is advantageously implemented N times in succession. Optionally, the probability of the fact to be found, such as the probability of whether a creature is located in an area, is obtained from the last partial result p_N. Advantageously, after the corresponding step, the data no longer needed is deleted from the working memory by means of a processor, so that the area of the working memory required for implementing the method is particularly small.
Claims
1. A method for operating a control device (SE), in particular a control device (SE) for detecting living beings or objects in an area, wherein: The control device (SE) has a processor (1) and a working memory (3), a first interface (7) for receiving measured values (x_i) and a second interface (9) for outputting results (E), wherein the method comprises the following steps: - providing the corresponding measured values (x_i) and the decision values (β_i) to a working memory (3), wherein a partial result (p_i) is calculated by means of a processor as a function of the corresponding decision value (β_i) and the measured value (x_i) and the partial result (p_i) is provided to the working memory (3), wherein the working memory (3) is designed to receive the corresponding decision value (β_i), the corresponding partial result (p_i) and the corresponding measured value (x_i); The following steps are implemented for i=1 to i=N respectively: - Calculate a further partial result (p_i+1) from a further measured value (x_i+1), a further decision value (β_i+1) and the partial result (p_i) in the corresponding preceding step (i); Therein, after i=N steps, the partial result (p_N) is provided as result (E) to the second interface (9).
2. The method according to claim 1, wherein: The calculation of the corresponding further partial result (p_i+1) takes place by multiplication of the measured value (x_i+1) and the corresponding decision value (β_i+1), wherein the product is added to the corresponding partial result (p_i) from the preceding step (i).
3. The method according to any one of the preceding claims, wherein: The decision values (β_i; i=1, . . . , N) are determined or have been determined by means of an algorithm with learning capability.
4. The method according to any one of the preceding claims, wherein: The method is implemented for evaluating measured values (x_i) of at least one sensor, in particular an ultra-wideband sensor.
5. The method according to any one of the preceding claims, wherein: After the calculation of the further partial result (p_i+1), the corresponding measured value (x_i), optionally the corresponding partial result (p_i) and the corresponding decision value (β_i) are removed from the working memory (3).
6. The method according to any one of the preceding claims, wherein: The control device (SE) has a further memory (5), wherein in a respective step (i) the respective decision value (β_i) is supplied from the further memory (5) to the working memory (3).
7. The method according to claim 6, wherein: The corresponding decision values (β_i) are stored in the form of a decision matrix in the further memory (5).
8. The method according to any one of the preceding claims, wherein: The presence or position of a living being in an area, in particular in a passenger compartment of a vehicle, is determined with the aid of a control device (SE).
9. The method according to any one of the preceding claims, wherein: The calculation is carried out in that in one step a plurality of measured values and a plurality of decision values (β_i) are provided to a working memory (3) and corresponding partial results (p_i) are provided for the respective plurality of measured values (x_i) and the plurality of decision values (β_i).
10. A control device (SE) comprising a processor (1) and a working memory (3) and optionally a further memory (5), further comprising a first interface (7) for receiving measured values (x_i) and a second interface (9) for outputting results (E), wherein: The control device (SE) is set up and designed to carry out a method according to one of the preceding claims.
11. A vehicle having a control device (SE) according to the preceding claim.