CN113296952A - System and method for performing edge calculation by adopting high-order differential of analog sensor - Google Patents
System and method for performing edge calculation by adopting high-order differential of analog sensor Download PDFInfo
- Publication number
- CN113296952A CN113296952A CN202110607304.4A CN202110607304A CN113296952A CN 113296952 A CN113296952 A CN 113296952A CN 202110607304 A CN202110607304 A CN 202110607304A CN 113296952 A CN113296952 A CN 113296952A
- Authority
- CN
- China
- Prior art keywords
- module
- sensor
- analog
- differential
- edge calculation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000004364 calculation method Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000004069 differentiation Effects 0.000 claims description 18
- 230000004044 response Effects 0.000 claims description 11
- 230000001133 acceleration Effects 0.000 claims description 5
- 230000003287 optical effect Effects 0.000 claims description 3
- 230000005855 radiation Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 abstract 1
- 238000012545 processing Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 7
- 238000003860 storage Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 239000012636 effector Substances 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 239000003990 capacitor Substances 0.000 description 3
- 230000014759 maintenance of location Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 230000026676 system process Effects 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 229910044991 metal oxide Inorganic materials 0.000 description 1
- 150000004706 metal oxides Chemical class 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000005293 physical law Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5072—Grid computing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention discloses a system for performing edge calculation by adopting high-order differential of an analog sensor, which comprises a sensor module, a differential module, an analog weighted summation module and a post-stage system module, wherein the sensor module is used for carrying out edge calculation on a sample; the four modules are connected in sequence; the sensor module collects input signals and generates electrical analog signals; the differential module obtains a differential signal of the electrical analog signal; the analog weighted summation module is used for weighting and summing differential signals of each order; and the rear-stage system module controls a rear-stage system according to the output of the analog weighted summation module. The invention also discloses a method for performing edge calculation by adopting the high-order differential of the analog sensor. The method realizes Taylor expansion of continuous input signals in a hardware circuit, and carries out preprocessing on motion signals collected by a sensor so as to predict the motion of an object; the invention reduces the power consumption of the system, reduces the time required by calculation, improves the endurance time of the system, reduces the system delay and improves the accuracy and the performance of the system.
Description
Technical Field
The present invention relates to a system and a method for edge calculation, and more particularly, to a system and a method for edge calculation using analog sensor high-order differentiation.
Background
The sensors are a critical loop in an automatic control system. The existing sensor device can sense corresponding environment information and convert the environment information into electric signals or output information in other available forms according to a certain rule so as to meet the requirements of recording, transmitting, storing, processing, controlling and the like of information in a control system. In many scenarios, intelligent pre-judgment needs to be performed on the acquired signals, and the most possible trend of the acquired signals in a future period of time is judged so as to make response preparation in advance. The method is of great significance particularly in control systems with large response delay, such as automatic braking technology for driving assistance, target capturing and tracking technology and the like.
The current technologies related to signal prediction, such as track prediction, target capture and tracking, are mainly realized by a hardware architecture consisting of a sensor unit and a processing unit. The control system acquires the environmental information through various sensors, and after the environmental information is digitally sampled and processed through a digital control circuit taking a digital processor as a core, the development trend of signals is predicted, so that the control system makes corresponding response to the environment and the information. The core of the processing unit is mainly designed with a digital controller circuit and related algorithms driving intelligent prediction applications. The integrated digital chip is a logic circuit based on complementary metal oxide semiconductor technology and made of transistors as basic elements, and the main components of the integrated digital chip are circuit modules such as a general processor, a memory and the like. One of the major bottlenecks that limit performance when performing computations is the limited speed of information exchange between memory and the processor. The sensing unit is composed of sensors which are made by different corresponding device principles for different physical quantities and changes thereof. This response is electrically an analog signal. In order to match with a digital processing system, after the sensor element converts the analog quantity in the environment into the electrical analog quantity, the sensor element also needs to be matched with an analog-to-digital converter to convert the analog quantity into a digital signal, and a large amount of energy is consumed in the conversion process, and information is lost. The digital information is communicated to a controller circuit of a later stage for subsequent signal communication and calculation processing. In the process of calculation processing, algorithms used in cooperation with a digital calculation system all face contradictions between precision and calculation time, and if high-precision calculation is required to be met, the calculation amount and the calculation time are increased sharply, so that the system is difficult to complete real-time processing.
The unmanned aerial vehicle project of the zurich rational team and the Vijay Kumar team shows that the unmanned aerial vehicle smoothly captures and rebounds an object thrown into the air. The information of the environment and the position change of the object is sensed by a large-scale sensor system, and the calculation is carried out through a series of algorithms, so that the position and the speed of the object at the next moment are pre-judged, and finally, the target function can be realized. However, the information processing of the project is a means based on computer programming, and a calculation and processing algorithm is executed on a digital circuit, so that the complexity and the power consumption are greatly inferior to those of the scheme.
Due to the division of the sensor and the processor in the analog domain and the digital domain, and the need of a separate processing unit architecture for the digital processing chip, the exchange and processing speed of the information is often greatly restricted, thereby causing a large delay between the response after the environment detection and the information processing. With the increase of information amount caused by the complexity of the environment, the pressure of sensor information on transmission bandwidth in the transmission process is increasing. In addition, the structural complexity of the system is relatively high due to the complex digital circuits required to perform specific arithmetic calculations and the conversion between the sensor analog signal and the digital signal and between the digital signal and the control analog signal, which results in high cost.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a system and a method for performing edge calculation by adopting high-order differential of an analog sensor, which solve the problems of high delay, high cost and the like in the prior art.
The technical scheme is as follows: the invention relates to a system for performing edge calculation by adopting high-order differential of an analog sensor, which comprises a sensor module, a differential module, an analog weighted summation module and a post-stage system module; the output of the sensor module is connected with the input of the differential module, the output of the differential module is connected with the input of the analog weighted summation module, and the output of the analog weighted summation module is connected with the input of the post-stage system module; the sensor module collects input signals and generates electrical analog signals f (t); the differential module obtains 1, 2 … n-order differential signals of an input electrical analog signal f (t), wherein n is a positive integer; the analog weighted summation module is used for weighted summation of f (t) differential signals of various orders; and the rear-stage system module controls a rear-stage system according to the output of the analog weighted summation module.
The sensor module includes any kind and any number of sensors. The sensor comprises at least one of an optical sensor, an electrical sensor, a pressure sensor, a temperature sensor, a sound sensor, a speed sensor, an acceleration sensor, a radiation sensor, a heat-sensitive sensor and a magnetic-sensitive sensor.
The differential module is formed by cascading N differential circuits, the output of the previous differential circuit is used as the input of the next differential circuit, and N is larger than or equal to 2.
The weighted value of the analog weighted summation module is the input electrical analog signal f (t0 in Taylor expansion term
Wherein t is0And the current time is t, the predicted time is t, and n is an integer.
The analog weighted summation module is a memristor or an electronic element with nonvolatile signal modulation capability. The electronic element is one or more of MRAM, PCM, FLASH, constant value resistance and variable resistance.
The rear-stage system module is a control response unit or a motion control unit.
The invention relates to a method for performing edge calculation by adopting high-order differentiation of an analog sensor, which comprises the following steps of:
(1) collecting an input signal by using a sensor, and generating an electrical analog signal f (t);
(2) obtaining 1, 2 … n-order differential signals of the input electrical analog signals f (t) by adopting a differential module, wherein n is a positive integer;
(3) weighting and summing differential signals of each order by adopting an analog weighted summing module; weighted values in terms of Taylor expansion of the input electrical analogue signal f (t)
Wherein t is0Is the current time, t is the predicted time, and n is a positive integer.
(4) And controlling a post-stage system by using the output of the analog weighted summation module.
Has the advantages that: compared with the prior art, the method has the remarkable advantages that Taylor expansion of continuous input signals is realized on a hardware circuit, and motion signals collected by the sensor are preprocessed, so that the motion of an object is predicted; the invention reduces the power consumption of the system, reduces the time required by calculation, improves the endurance time of the system, reduces the system delay and improves the accuracy and the performance of the system.
Drawings
FIG. 1 is a block diagram of the system of the present invention;
FIG. 2 is a schematic diagram of a differential module of the present invention;
FIG. 3 is a schematic diagram of an analog weighted sum module of the present invention;
FIG. 4 is an embodiment of a differentiating circuit of the present invention;
FIG. 5 is another embodiment of a differentiating circuit of the present invention;
FIG. 6 is an embodiment of an analog weighted sum module of the present invention;
FIG. 7 is a schematic diagram of an embodiment of the invention for early warning of vehicle collision and early airbag activation;
fig. 8 is a schematic view of an embodiment of the airbag crash-proof safety suit according to the present invention.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
As can be seen from FIG. 1, the system for performing edge calculation by using analog sensor high-order differentiation according to the present invention includes a sensor module, a differentiation module, an analog weighted summation module, and a post-system module; the output of the sensor module is connected with the input of the differential module, the output of the differential module is connected with the input of the analog weighted summation module, and the output of the analog weighted summation module is connected with the input of the post-stage system module; the sensor module collects input signals and generates electrical analog signals f) t); the differential module obtains 1, 2 … n-order differential signals of an input electrical analog signal f (t), wherein n is a positive integer; the analog weighted summation module is used for weighting and summing the differential signals of each order of the electrical analog signals f (t); and the rear-stage system module controls a rear-stage system according to the output of the analog weighted summation module.
The detection signal of the sensor module is used as input and is subjected to differential processing through a differential module. The analog weighted summation module which modulates according to different prediction time combines the differential signals, and the obtained output result can reflect the prediction of the system on the change trend of the analog signals, so that an effector of a later-stage system module can be guided to generate corresponding response.
The sensors used in the present invention may be any kind and number of sensors capable of continuously detecting a changing environment, such as optical, electrical, pressure, temperature, sound sensors, speed sensors, acceleration sensors, radiation sensors, thermal sensors, magnetic sensors, etc. Different sensors are adopted according to different specific requirements.
The differential module adopted by the invention is realized by cascade of differential circuits. Differential circuits of different structures and devices differ in accuracy, complexity and maximum speed. In practical application, different differentiating circuits can be selected according to different practical requirements. By further differentiating the input signal and further inputting the calculated differentiated signal to another differentiating circuit, a series of high-order differentiated input signals can be obtained. In general, signal orders greater than 5 can be used to obtain better results. As shown in fig. 2, a series of high-order differential input signals can be obtained by performing a differential process on the input signal and further differentiating the corresponding differential signals.
The analog weighted summation module is shown in fig. 3, and performs weighted summation on the series of high-order differential signals obtained in the foregoing manner in an analog domain according to a certain weight. The continuous sensor input signal is regarded as a differentiable function f (t) with respect to time, for which x is0Performing Taylor expansion
Wherein the content of the first and second substances,
the term is defined by the weight in this module. T-t ═ t0I.e. the time interval between the current time and the predicted time. The module takes the currently input high-order differential signal as an input vector, and performs dot product summation with given weight in an analog circuit, and the result can be used for representing the prediction output after the time delta t. By setting
The magnitude of the term may control the time difference between the predicted signal and the current signal.
Generally, the components used as the weight values in the analog weighted sum module may be memristors, or may include nonvolatile electronic components with signal modulation capability, such as MRAM, PCM, FLASH, fixed-value resistor, and variable resistor. The selection of these specific devices does not change the design core concept of the present invention. The storages are combined into a cross array, and random storages are arranged at the cross points. The memory can store the weight value w in proportion to
Input signal is proportional to
After flowing into the storage, the output signal y is proportional to f (t), then the analog signals of the outputs of several storages are added,
the latter system module is a control response module, and the specific selection is determined by actual requirements, such as a motion control system, a motor, a brake system and the like, a latter processor and the like. And different functions and applications can be realized according to the prediction result output by the front-stage module by matching with different rear-stage systems.
In this embodiment, the differentiation module is based on an operational amplifier, as shown in fig. 4. The differential circuit can realize accurate differentiation of input signals under ideal conditions by connecting a proper capacitor with an inverse amplifying circuit consisting of an operational amplifier in series. The structure has the characteristic of high precision and is suitable for being applied under the condition of high precision requirement. As shown in fig. 5, the differentiation module may also be based on an RC circuit. Under the conditions of limited cost requirement, low precision requirement or extremely high signal processing speed requirement, the capacitor and the resistor are directly connected in series, and approximate high-order differential signals can be obtained at the connection position of the capacitor and the resistor.
In this embodiment, the analog weighted sum module is a weighted sum module based on a memristor array, as shown in fig. 6. Wherein the input signal is composed of original signal f (t), first order differential signal f1And a higher order differential signal f2、f3…fnAnd (4) forming. The resistance value of the memory corresponding to each differential signal represents the weight value corresponding to the original signal after Taylor expansion
Assuming the system is used to predict the sensor signal one second later in the current environment, (t-t)0)n1, the resistance value of the storage corresponding to each order of differential signal is
The proportional relationship of (c). When the system sets the predicted time delta t as t-t0When the resistance value proportion of the corresponding storage is changed, the resistance value proportion of the corresponding storage is correspondingly changed. This can be achieved by adjusting the memory in the weighted sum module and can be processed in parallel, resulting in predicted signal outputs I at different times1、I2、I3。
A method for performing edge calculation using analog sensor higher order differentiation, comprising the steps of:
(1) collecting an input signal by using a sensor, and generating an electrical analog signal f (t);
(2) obtaining 1, 2 … n-order differential signals of the input electrical analog signals f (t) by adopting a differential module, wherein n is a positive integer;
(3) weighting and summing differential signals of each order by adopting an analog weighted summing module; weighted values in terms of Taylor expansion of the input electrical analogue signal f (t)
Wherein t is0Is the current time, t is the predicted time, and n is a positive integer.
(4) And controlling a post-stage system by using the output of the analog weighted summation module.
In this embodiment, the sensor module is a vehicle-mounted sensor, a laser range radar, a camera, and the like. The rear-stage system module is a vehicle safety system for opening the air bag. As shown in fig. 7. The edge calculation module 3 of the invention is connected with a vehicle-mounted sensor laser ranging radar 2, a camera and the like, a rear-end effector brake and an air bag 1. In the running process of the vehicle, input data acquired by a laser radar or a camera is preprocessed and predicted by the high-order differential module, and potential driving risks are analyzed.
In the running process of the vehicle, the original input signal is the distance between the vehicle and the obstacle in front of the vehicle, the pedestrian and the like which are constantly changed. When the system processes the input signal, the change of the signal exceeding the threshold value is predicted, for example, in the case that the distance of the obstacle is suddenly reduced, namely collision occurs, the system sends a signal to control the braking system to perform emergency braking, so that the collision is prevented.
In addition, the system can realize the early deployment of the inflatable airbag by means of changing threshold values, effectors and the like. Therefore, the airbag can be fully deployed in time under the condition that collision cannot be avoided, and the safety of a driver and passengers is protected. Because the processing part of the system is based on hardware and physical laws, and the processing and response speed has the advantage of magnitude compared with a digital-based system, the system can make a correct response in a shorter time and realize a safer driving scheme.
Meanwhile, the sensor signal is directly processed at the front end, so that the pressure of the information transmission process on the bandwidth can be reduced. Meanwhile, the system structure and the selectable hardware architecture which are concise can greatly reduce the cost required for realizing the similar functions because the digital circuit is not required to support.
The rear-stage system module can also be an anti-falling safety suit of the air bag. As shown in fig. 8, the edge calculation module 3 of the present invention connects a height sensor and an acceleration sensor on the lap belt of the safety garment with the rear end effector airbag 1. In the working process, the data acquired by the height sensor and the acceleration sensor are preprocessed by the high-order differential module of the invention, and the potential risks of the user are analyzed. The target group of the anti-falling safety suit can be a group with a large potential collision risk, such as the old or a motorcycle driver who easily falls down at home.
When the system processes the input signal, and predicts that the signal exceeds a threshold value, for example, the old man wearing the anti-collision safety suit is predicted to fall down soon according to the condition that the height is small and is about to be 0 and the speed is increased sharply, or the motorcycle driver wearing the anti-collision safety suit is predicted to collide according to the speed change, a signal is sent out to trigger an air bag on the anti-collision safety suit, so that the air bag can be used for wrapping the human body in a very short time, and the injury to the human body caused by collision is prevented to the maximum extent.
According to the invention, through a set of differential circuit module and analog weighting summation circuit module, high-order differential derivation is carried out on continuous sensor analog signals, Taylor series is calculated in an analog domain, and position information and speed information of an object at the next moment are judged with controllable precision. Various functions can be realized by matching with other different peripheral circuits, such as an effector module.
Theoretically this method has the ability to accurately approximate arbitrary functions by adding higher order differential terms. Therefore, for the continuously changing external environment, the invention can preprocess the input signal at the input end of the sensor to directly obtain the prediction result of the state at the moment in the environment, and the obtained result is accurate on a shorter time scale.
Due to the characteristics of low structural complexity and high operation speed, the delay caused by processing input data in the prior art can be reduced. Compared with the prior art which can realize the functions, the invention has simple structure, lower required cost and huge potential and advantages for commercial application.
Claims (10)
1. A system for performing edge calculation using analog sensor higher order differentiation, comprising: the system comprises a sensor module, a differential module, an analog weighted summation module and a post-stage system module; the output of the sensor module is connected with the input of the differential module, the output of the differential module is connected with the input of the analog weighted summation module, and the output of the analog weighted summation module is connected with the input of the post-stage system module;
the sensor module collects input signals and generates electrical analog signals f (t);
the differential module obtains 1, 2 … n-order differential signals of the electrical analog signals f (t), wherein n is a positive integer;
the analog weighted summation module is used for weighted summation of differential signals of each order of f (t);
and the rear-stage system module controls a rear-stage system according to the output of the analog weighted summation module.
2. The system for edge calculation using analog sensor higher order differentiation according to claim 1, wherein: the sensor module includes any kind and any number of sensors.
3. The system for edge calculation using analog sensor higher order differentiation according to claim 2, wherein: the sensor comprises at least one of an optical sensor, an electrical sensor, a pressure sensor, a temperature sensor, a sound sensor, a speed sensor, an acceleration sensor, a ray radiation sensor, a heat-sensitive sensor and a magnetic-sensitive sensor.
4. The system for edge calculation using analog sensor higher order differentiation according to claim 1, wherein: the differential module comprises N cascaded differential circuits, the output of the previous differential circuit is used as the input of the next differential circuit, and N is larger than or equal to 2.
5. The system for edge calculation using analog sensor higher order differentiation according to claim 1, wherein: the weighted value of the analog weighted summation module is in the Taylor expansion term of the input electrical analog signal f (t)
Wherein t is0Is the current time, t is the predicted time, and n is a positive integer.
6. The system for edge calculation using analog sensor higher order differentiation according to claim 1, wherein: the analog weighted sum module includes memristors or electronic components with non-volatile signal modulation capability.
7. The system for edge calculation using analog sensor higher order differentiation according to claim 6, wherein: the electronic element comprises one or more of MRAM, PCM, FLASH, fixed-value resistance and variable resistance.
8. The system for edge calculation using analog sensor higher order differentiation according to claim 1, wherein: the post-stage system module comprises a control response unit or a motion control unit.
9. A method for performing edge calculation by adopting analog sensor high-order differentiation is characterized by comprising the following steps: the method comprises the following steps:
(1) collecting an input signal by using a sensor, and generating an electrical analog signal f (t);
(2) obtaining 1, 2 … n-order differential signals of the input electrical analog signals f (t) by adopting a differential module, wherein n is a positive integer;
(3) weighting and summing the differential signals of each order of f (t) by adopting an analog weighted summing module;
(4) and controlling a post-stage system by using the output of the analog weighted summation module.
10. The method for performing edge calculation using analog sensor higher order differentiation according to claim 9, wherein: the weighting value in step (3) is the value in Taylor expansion term of the input electrical analog signal f (t)
Wherein t is0Is the current time, t is the predicted time, and n is a positive integer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110607304.4A CN113296952B (en) | 2021-06-01 | 2021-06-01 | System and method for performing edge calculation by adopting high-order differential of analog sensor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110607304.4A CN113296952B (en) | 2021-06-01 | 2021-06-01 | System and method for performing edge calculation by adopting high-order differential of analog sensor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113296952A true CN113296952A (en) | 2021-08-24 |
| CN113296952B CN113296952B (en) | 2022-03-18 |
Family
ID=77326661
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110607304.4A Active CN113296952B (en) | 2021-06-01 | 2021-06-01 | System and method for performing edge calculation by adopting high-order differential of analog sensor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113296952B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1917576A (en) * | 2006-08-30 | 2007-02-21 | 蒲亦非 | Fractional order differential filter for digital image |
| CN102902274A (en) * | 2012-08-08 | 2013-01-30 | 空军工程大学航空航天工程学院 | Self-adaptive weighting differential game guidance method |
| CN203596803U (en) * | 2013-11-25 | 2014-05-14 | 武汉理工大学 | Frequency compensation device |
| CN106353788A (en) * | 2016-09-29 | 2017-01-25 | 成都理工大学 | Analog digital mixed pulse amplitude analyzer and analysis technology thereof |
| CN107707259A (en) * | 2017-11-01 | 2018-02-16 | 兰州大学 | A kind of method of analog signal sampling and reconstruct |
| CN110034968A (en) * | 2019-03-12 | 2019-07-19 | 上海交通大学 | Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations |
| CN110842915A (en) * | 2019-10-18 | 2020-02-28 | 南京大学 | Robot control system and method based on memristor cross array |
| CN111189553A (en) * | 2020-01-10 | 2020-05-22 | 北京航天测控技术有限公司 | Thermocouple and synchronous acquisition device for multi-order differential signals thereof |
| CN111276125A (en) * | 2020-02-11 | 2020-06-12 | 华南师范大学 | Lightweight speech keyword recognition method facing edge calculation |
| CN213072581U (en) * | 2020-09-28 | 2021-04-27 | 杭州龙康电子有限公司 | Reliable sensor analog amplification circuit |
-
2021
- 2021-06-01 CN CN202110607304.4A patent/CN113296952B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1917576A (en) * | 2006-08-30 | 2007-02-21 | 蒲亦非 | Fractional order differential filter for digital image |
| CN102902274A (en) * | 2012-08-08 | 2013-01-30 | 空军工程大学航空航天工程学院 | Self-adaptive weighting differential game guidance method |
| CN203596803U (en) * | 2013-11-25 | 2014-05-14 | 武汉理工大学 | Frequency compensation device |
| CN106353788A (en) * | 2016-09-29 | 2017-01-25 | 成都理工大学 | Analog digital mixed pulse amplitude analyzer and analysis technology thereof |
| CN107707259A (en) * | 2017-11-01 | 2018-02-16 | 兰州大学 | A kind of method of analog signal sampling and reconstruct |
| CN110034968A (en) * | 2019-03-12 | 2019-07-19 | 上海交通大学 | Multi-sensor Fusion vehicle safety method for detecting abnormality based on edge calculations |
| CN110842915A (en) * | 2019-10-18 | 2020-02-28 | 南京大学 | Robot control system and method based on memristor cross array |
| CN111189553A (en) * | 2020-01-10 | 2020-05-22 | 北京航天测控技术有限公司 | Thermocouple and synchronous acquisition device for multi-order differential signals thereof |
| CN111276125A (en) * | 2020-02-11 | 2020-06-12 | 华南师范大学 | Lightweight speech keyword recognition method facing edge calculation |
| CN213072581U (en) * | 2020-09-28 | 2021-04-27 | 杭州龙康电子有限公司 | Reliable sensor analog amplification circuit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113296952B (en) | 2022-03-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Real-time human motion behavior detection via CNN using mmWave radar | |
| Ulrich et al. | Deepreflecs: Deep learning for automotive object classification with radar reflections | |
| US4320287A (en) | Target vehicle tracking apparatus | |
| AU2021240237A1 (en) | Trajectory prediction method and device | |
| US8204927B1 (en) | System and method for cognitive processing for data fusion | |
| Rosenband | Inside Waymo's self-driving car: My favorite transistors | |
| CN113296952B (en) | System and method for performing edge calculation by adopting high-order differential of analog sensor | |
| US20200278423A1 (en) | Removing false alarms at the beamforming stage for sensing radars using a deep neural network | |
| Dbouk et al. | KeyRAM: A 0.34 uJ/decision 18 k decisions/s recurrent attention in-memory processor for keyword spotting | |
| Lee et al. | 14.2 a 502gops and 0.984 mw dual-mode adas soc with rnn-fis engine for intention prediction in automotive black-box system | |
| KR20220054659A (en) | Drive aids, methods, vehicles and storage media | |
| CN111076770A (en) | Multi-mode intelligent sensor with sensing element and memristor combined | |
| Chen et al. | Triboelectric nanogenerator sensors for intelligent steering wheel aiming at automated driving | |
| EP3944049B1 (en) | Mobile communication terminal device operation of robot terminal | |
| Swamy et al. | FPGA accelerated automotive ADAS sensor fusion | |
| CN106740874A (en) | A kind of intelligent travelling crane early warning sensory perceptual system based on polycaryon processor | |
| CN102472999A (en) | Control target processing system | |
| CN116359846A (en) | Dynamic millimeter wave Lei Dadian cloud human body analysis method based on joint learning | |
| CN114861902A (en) | Processing unit, operation method thereof and computing chip | |
| Varagula et al. | Object detection method in traffic by on-board computer vision with time delay neural network | |
| CN110654339B (en) | Method and device for early detection of accidents | |
| Yuan et al. | A centralised training algorithm with D3QN for scalable regular unmanned ground vehicle formation maintenance | |
| CN220708749U (en) | Equipment test system for test vehicle and test vehicle thereof | |
| Wang et al. | Research on gesture recognition algorithm based on millimeter-wave radar in vehicle scene | |
| Holzbock et al. | Gesture recognition with keypoint and radar stream fusion for automated vehicles |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |