WO2022241620A1 - 用于交通工具的集成电路芯片 - Google Patents

用于交通工具的集成电路芯片 Download PDF

Info

Publication number
WO2022241620A1
WO2022241620A1 PCT/CN2021/094166 CN2021094166W WO2022241620A1 WO 2022241620 A1 WO2022241620 A1 WO 2022241620A1 CN 2021094166 W CN2021094166 W CN 2021094166W WO 2022241620 A1 WO2022241620 A1 WO 2022241620A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
bus
circuit
coupled
sensory
Prior art date
Application number
PCT/CN2021/094166
Other languages
English (en)
French (fr)
Inventor
夏晶
段楠
蔡春晓
信恒超
Original Assignee
华为技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2021/094166 priority Critical patent/WO2022241620A1/zh
Priority to CN202180093275.2A priority patent/CN116888927A/zh
Priority to EP21940077.7A priority patent/EP4325789A4/en
Publication of WO2022241620A1 publication Critical patent/WO2022241620A1/zh
Priority to US18/510,972 priority patent/US20240089143A1/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/40—Bus networks
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/04—Programme control other than numerical control, i.e. in sequence controllers or logic controllers
    • G05B19/042—Programme control other than numerical control, i.e. in sequence controllers or logic controllers using digital processors
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B9/00—Safety arrangements
    • G05B9/02—Safety arrangements electric
    • G05B9/03—Safety arrangements electric with multiple-channel loop, i.e. redundant control systems
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/20—Pc systems
    • G05B2219/26—Pc applications
    • G05B2219/2637—Vehicle, car, auto, wheelchair
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/40—Bus networks
    • H04L2012/40267—Bus for use in transportation systems
    • H04L2012/40273—Bus for use in transportation systems the transportation system being a vehicle
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks

Definitions

  • the present disclosure relates to the field of electronics, and more particularly to integrated circuit chips and electronic components for vehicles.
  • embodiments of the present disclosure aim to provide a chip, an electronic component, a vehicle, and a method for generating a control signal for implementing control of the vehicle in a low-cost manner.
  • a chip includes a first bus, a second bus and a processor.
  • the first bus is configured to transmit first sensing data.
  • the second bus is coupled to the first bus.
  • the type of the second bus is different from the type of the first bus.
  • the second bus is configured to transmit the second sensory data and the first sensory data from the first bus.
  • the type of the second sensing data is different from the type of the first sensing data.
  • a processor is coupled to the second bus, and the processor is configured to generate a control signal for controlling the actuator based on the first sensory data and the second sensory data.
  • each functional circuit By coupling each functional circuit with different buses based on the function type (such as sensing and decision-making), the sensing data can be transmitted on the first bus, and the decision-making data can be transmitted on the second bus, so that the advantages of each bus can be effectively utilized , such as the low latency and efficient data transmission of the ring bus and the strong scalability, large bandwidth and low latency of the grid bus, and reduce resource consumption and delay.
  • the function type such as sensing and decision-making
  • the first bus is selected from the group consisting of a mesh bus, a fully connected bus and a star bus.
  • the second bus is selected from the group consisting of a ring bus, a linear bus and a tree bus.
  • the chip further includes a first processing circuit.
  • a first processing circuit is coupled to the first bus and receives first sensed data from a first sensor coupled to the first bus.
  • the first processing circuit processes the first sensing data to generate first sensing data.
  • the first processing circuit is selected from the group consisting of artificial intelligence vector circuits and artificial intelligence core circuits.
  • the first sensor is selected from the group consisting of camera, lidar, sonar and radar.
  • the chip further includes a third bus.
  • the third bus is coupled to the first bus.
  • the type of the third bus is the same as that of the second bus.
  • the third bus is configured to transmit first assistance data corresponding to the first sensory data and second assistance data corresponding to the second sensory data from the first bus.
  • the processor is coupled to the third bus and the processor is further configured to generate a control signal based on the first sensory data, the first assistance data, the second sensory data, and the second assistance data.
  • the auxiliary data includes cyclic redundancy check data.
  • the chip further includes a second processing circuit.
  • a second processing circuit is coupled to the first bus or the second bus and receives second sensed data from a second sensor coupled to the first bus or the second bus. The second processing circuit processes the second sensing data to generate second sensing data.
  • the second processing circuit is selected from the group consisting of artificial intelligence vector circuits and artificial intelligence core circuits.
  • the second sensor is selected from the group consisting of camera, lidar, sonar and radar.
  • the processor is further configured to determine, based on the first perception data and the first auxiliary data, first environmental data representing the environmental conditions; second environmental data of the condition; and generating a control signal based on the first environmental data and the second environmental data.
  • the processor is further configured to fuse the first environmental data and the second environmental data to generate fused data; and generate a control signal based on the fused data.
  • the processor is further configured to fuse the first sensing data and the second sensing data to generate fusion data; and generate a control signal based on the fusion data.
  • the chip further includes: a first cross-site circuit located on the first bus and configured to transmit data; and a first processing circuit coupled to the first cross-site circuit and configured to First sensing data from a sensor that is coupled to the first cross-site circuit generates first sensing data.
  • the chip further includes a second cross-site circuit on the second bus and configured to transmit data; and a second processing circuit coupled to the second cross-site circuit and configured to The second sensing data of the second sensor is coupled to the second cross-site circuit to generate second sensing data.
  • the first bus includes a mesh bus
  • the second bus includes a ring bus
  • the third bus includes a ring bus
  • an electronic component includes a circuit board and a chip according to the first aspect. Chips are mounted on circuit boards. By coupling each functional circuit in the chip to different buses based on the function type (such as sensing and decision-making), the sensing data can be transmitted on the first bus, and the decision-making data can be transmitted on the second bus, so that the various circuits can be effectively used respectively.
  • the advantages of the bus such as the low latency and efficient data transmission of the ring bus and the strong scalability, large bandwidth and low latency of the grid bus, and reduce resource consumption and delay.
  • a vehicle includes a power source; and an electronic assembly according to the second aspect.
  • the electronic components are powered by the power supply.
  • Each functional circuit in the chip in the vehicle is coupled to different buses based on the function type (such as sensing and decision-making), and the sensing data can be transmitted on the first bus, and the decision-making data can be transmitted on the second bus, so that they can be respectively Effectively utilize the advantages of each bus, such as the low latency and high-efficiency data transmission of the ring bus and the strong scalability, large bandwidth and low latency of the grid bus, and reduce resource consumption and delay.
  • the vehicle also includes a first sensor and a second sensor.
  • a first sensor is coupled to the first bus via a first interface and is configured to generate and transmit first sensor data to the first bus.
  • a second sensor is coupled to the second bus via the second interface and is configured to generate and transmit second sensor data to the second bus.
  • a method for generating a control signal includes transmitting the first sensory data through the first bus; transmitting the second sensory data and the first sensory data from the first bus through the second bus, the type of the second bus is different from the type of the first bus, and the second sensory The type of the data is different from the type of the first sensing data; a control signal for controlling the actuator is generated based on the first sensing data and the second sensing data acquired from the second bus.
  • Each functional circuit in the chip is coupled to different buses based on the function type (such as sensing and decision-making), so that sensing data can be transmitted on the first bus, and decision-making data can be transmitted on the second bus, so that each bus can be effectively used respectively Advantages, such as low latency and high-efficiency data transmission of the ring bus and strong scalability, large bandwidth and low latency of the grid bus, and reduce resource consumption and delay.
  • function type such as sensing and decision-making
  • the method further includes transmitting the first auxiliary data corresponding to the first sensory data and the second auxiliary data corresponding to the second sensory data from the first bus through a third bus, the third bus being coupled to the first a bus, a third bus of the same type as the second bus; and generating a control signal based on the first sensing data, the first auxiliary data, the second sensing data, and the second auxiliary data.
  • the third bus of the same type as the second bus to transmit auxiliary data, it can be used to verify the integrity or correctness of the sensing data transmitted on the second bus to improve the security of data transmission, thereby improving the security performance of the chip .
  • generating the control signal for controlling the execution device based on the first sensing data and the second sensing data obtained from the second bus includes determining a signal representing the environmental condition based on the first sensing data and the first auxiliary data.
  • first environmental data determining second environmental data representing an environmental condition based on the second sensory data and the second auxiliary data; and generating a control signal based on the first environmental data and the second environmental data.
  • the correctness of each other's environmental data can be verified interactively, for example, if the first environmental data indicates that there are pedestrians on the road ahead, and the second environmental data of different types also indicates that there are pedestrians on the road ahead, then it can be based on consistency The result generates the corresponding control signal.
  • generating the control signal based on the first environment data and the second environment data includes fusing the first environment data and the second environment data to generate fusion data; and generating the control signal based on the fusion data.
  • the correctness of each other's environmental data can be verified interactively, for example, if the first environmental data indicates that there are pedestrians on the road ahead, and the second environmental data of different types also indicates that there are pedestrians on the road ahead, then it can be based on consistency The result generates the corresponding control signal.
  • FIG. 1 shows a schematic diagram of an environment according to some embodiments of the present disclosure
  • Figure 2 shows a schematic block diagram of an electronic assembly according to some embodiments of the present disclosure
  • Figure 3 shows a schematic diagram of a chip according to some embodiments of the present disclosure
  • Fig. 4 shows a schematic diagram of a ring bus cross station according to some embodiments of the present disclosure
  • Fig. 5 shows a schematic diagram of a mesh bus cross-site according to some embodiments of the present disclosure
  • Fig. 6 shows a schematic diagram of data transmission via a mesh bus according to some embodiments of the present disclosure
  • FIG. 7 shows a schematic diagram of data transmission via a ring bus according to some embodiments of the present disclosure.
  • Fig. 8 shows a schematic flowchart of a method for generating a control signal according to some embodiments of the present disclosure.
  • the term “comprising” and its similar expressions should be interpreted as an open inclusion, that is, “including but not limited to”.
  • the term “based on” should be understood as “based at least in part on”.
  • the term “one embodiment” or “the embodiment” should be read as “at least one embodiment”.
  • the terms “first”, “second”, etc. may refer to different or the same object.
  • the term “and/or” means at least one of the two items associated with it. For example "A and/or B" means A, B, or A and B. Other definitions, both express and implied, may also be included below.
  • the security performance of chips used in vehicles is especially important.
  • the ISO 26262 standard divides functional safety into four safety levels (Automotive Safety Integrity Level, ASIL) of A, B, C, and D, among which ASIL-D is the highest level and represents the most stringent safety requirements.
  • ASIL Automotive Safety Integrity Level
  • AIC artificial intelligence core
  • Each chip usually integrates many functional circuits, such as processors, memories, buffers, artificial intelligence vector (AIV) circuits, artificial intelligence core (AIC) circuits, interface circuits, etc., each of which There may be one or more functional circuits.
  • These functional circuits are respectively coupled to multiple cross station (cross station, CS) circuits on the bus, and the multiple CS circuits can be interconnected through the bus.
  • cross station cross station
  • the CS circuits on the bus receive and send data in a beating manner according to the frequency.
  • a CS circuit can receive data from a bus via the bus and transfer that data to a functional circuit connected to it in a beating manner, or transfer data from a functional circuit to other functions on the bus in a beating manner. circuit.
  • the CS circuit can relay data on the bus in a beating fashion.
  • the CS circuit enables various functional circuits to transfer instructions and data between each other via the bus. For architectures that use a dual-chip redundant design, this requires each chip to have essentially the same configuration to achieve redundant identical functions.
  • each chip usually has the same type of main bus and sub bus, such as a mesh bus, to ensure the security of data between each circuit in each chip, for example, to ensure that data transmission within the chip does not will be lost or generate an error. Therefore, in some conventional schemes, there is actually a two-chip architecture with the same type of four buses to ensure security.
  • main bus and sub bus such as a mesh bus
  • One conventional bus type is a ring bus, on which each CS circuit can couple 2 functional circuits.
  • the ring bus has low delay and fast and efficient data transmission.
  • the transmission delay between the various functional circuits becomes larger.
  • the carrying bandwidth of the ring bus is limited, when there is a large amount of data transmission demand between various functional circuits, the traffic congestion of the ring circuit will be caused and the bandwidth will be limited.
  • Another conventional bus type is a mesh bus, and each CS circuit on the mesh bus can couple four functional circuits.
  • the mesh bus is a two-dimensional bus network formed by crossing multiple horizontal ring buses and multiple vertical ring buses.
  • the mesh bus has strong scalability, and the functional circuits on the mesh bus can have large bandwidth and low delay between each other.
  • the structure of the mesh bus is too complicated, consumes a lot of resources and has a large delay.
  • a dual-chip redundant design is often used in conventional designs, this will result in greater cost and resource consumption.
  • each functional circuit is combined with a first bus such as a mesh bus and a second bus such as a ring bus within a single chip based on the type of function (such as sensing and decision-making). Coupling, the perception data can be transmitted on the first bus, and the decision data can be transmitted on the second bus, so that the advantages of each bus can be effectively utilized, such as the low latency and high-efficiency data transmission of the ring bus and the scalability of the grid bus Strong, large bandwidth and low latency, and reduce resource consumption and latency.
  • sensing data such as camera data and LiDAR data
  • it is possible to verify the correctness of the sensing data ensure that the correct environmental objects are sensed and generate corresponding decision-making or control signals, so that traffic Actuating devices of the tool, such as actuators, can make a corresponding correct operation.
  • traffic Actuating devices of the tool such as actuators
  • the safety level of the vehicle can be improved, for example reaching ASIL-D level.
  • Fig. 1 shows a schematic diagram of an environment according to some embodiments of the present disclosure.
  • a vehicle 100 includes an electronic assembly 20 .
  • the vehicle 100 is, for example, a car.
  • the vehicle 100 may also be other vehicles, such as motorcycles, air vehicles, or water vehicles.
  • Vehicle 100 also includes other components, such as various sensors and actuators.
  • Various sensors and actuators may be electrically coupled to electronics assembly 20 .
  • the sensors may, for example, include a camera, LiDAR, sorna, radar, temperature sensor, humidity sensor, etc., to sense environmental data and transmit the sensed data to the chip in the electronic component 20 .
  • the chips in the electronic component 20 analyze, compute and process various received data and generate decision signals or control signals, such as vehicle start signals, brake signals, turn signals, and air-conditioning signals.
  • Actuating devices such as actuators, brakes, and vehicle air conditioners can receive decision signals or control signals from the chips in the electronic assembly 20 to perform various operations, such as starting, braking, steering, and turning on the air conditioner.
  • FIG. 2 shows a schematic block diagram of an electronic assembly 20 according to some embodiments of the present disclosure.
  • the electronic assembly 20 includes a circuit board 21 and a plurality of electronic components mounted on the circuit board.
  • the circuit board 21 includes at least one of a printed circuit board (printed circuit board, PCB) and a flexible circuit board (flexbile printed circuit, FPC).
  • Multiple electronic components include a communication chip 22, a matching circuit 24, a power management chip 26 and a chip 30, wherein the communication chip 22 is used to communicate with external devices or other electronic devices, and the matching circuit 24 is used to match each electronic component
  • a power management chip 26 may, for example, be coupled to a power supply to manage received power and distribute power to various electronic components.
  • the chip 30 is used to receive sensing data from the above sensors to generate sensing data and generate control signals based on the sensing data.
  • Sensing data refers to raw data detected by sensors representing environmental conditions
  • sensing data refers to processed data representing or describing environmental conditions after processing the sensing data.
  • the original image or video data captured by the camera is sensing data, but the data that processes the image or video and recognizes and judges the presence of pedestrians in the captured image or video through AI algorithm and characterizes or describes it is perception data.
  • four electronic components are shown in FIG. 2 , it is understood that the electronic assembly 20 may include more or fewer electronic components without limitation of the present disclosure.
  • FIG. 3 shows a schematic diagram of a chip 30 according to some embodiments of the present disclosure.
  • the chip 30 includes a first bus M1, a second bus R1 and a third bus R2.
  • the chip 30 also includes a plurality of first CS circuits on the first bus M1, a plurality of second CS circuits on the second bus R1, and a plurality of third CS circuits on the third bus R2.
  • the chip 30 also includes a bridge circuit B1 coupled between the first bus M1 and the second bus R1, and a bridge circuit B2 coupled between the first bus M1 and the third bus R2.
  • the chip 30 also includes a first node circuit 371 , a second node circuit 372 and other circuits not shown. Although a specific configuration of the chip 30 is shown in FIG.
  • the chip 30 may include more or less circuits, and the present disclosure is not limited thereto.
  • a bridge circuit is used to couple different buses, the present disclosure is not limited thereto.
  • Other ways for coupling different buses are possible.
  • some functional circuits such as memory act as coupling circuits.
  • the sensing data on the first bus M1 is stored to the memory, and the processor on the second bus R1 retrieves the sensing data from the memory.
  • the first bus M1 is, for example, a mesh bus.
  • the first bus M1 is formed by, for example, 3 horizontal ring buses and 4 vertical ring buses intersecting each other, and a CS circuit is provided at each intersection point.
  • CS circuits 311, 312, 313 and 314 are provided at the intersections of the first horizontal ring bus and the four vertical ring buses
  • a CS circuit 311 is provided at the intersections of the first vertical ring bus and the three horizontal ring buses. , 321 and 331.
  • a specific mesh bus configuration is shown in 3 ⁇ 4 in FIG. 3 , this is only for illustration rather than limiting the scope of the present disclosure.
  • the horizontal buses and the vertical buses may have greater or fewer numbers, and the number of horizontal buses and the number of vertical buses may be the same or different.
  • the first bus M1 may also be another type of bus, such as a fully connected bus.
  • Each CS circuit on the first bus M1 is basically the same, so only a single CS circuit will be described below. It is understood that the description of this single CS circuit can be applied to other CS circuits on the first bus M1.
  • the first bus M1 is a mesh bus
  • each CS circuit may be coupled to four node circuits. Node circuits represent different functional circuits that may be coupled with CS circuits on the bus, eg, AIC circuits, AIV circuits, interface circuits, buffers, and so on. This disclosure does not impose any limitations on the nodal circuitry. Some external electronic devices, such as LiDAR, camera, sorna, etc., can be coupled to the CS circuit via an interface circuit.
  • each functional circuit is coupled on the bus by using a CS circuit, the present disclosure is not limited thereto. Other ways of coupling the various functional circuits to the bus are possible.
  • the vehicle 100 usually has multiple environmental condition sensors to sense the environmental conditions around the vehicle from multiple dimensions.
  • vehicle 100 may have 12 cameras, 4 LiDARs, 12 radars, and 4 sonars.
  • each interface circuit corresponding to the aforementioned plurality of sensors may be coupled to each CS circuit on the mesh circuit.
  • Chip 30 may have a plurality of processing circuits, such as AIV and/or AIC, respectively coupled to a plurality of CS circuits located on the first bus M1. Compared with the ring bus, this can avoid the congestion and limited bandwidth that the ring bus is prone to when multiple access circuits are involved.
  • Real-time sensed data can be stored in memory, and then read from memory by processing circuits such as AIV and/or AIC for preliminary processing to generate perception data representative of environmental conditions.
  • processing circuits such as AIV and/or AIC for preliminary processing to generate perception data representative of environmental conditions.
  • multiple processing circuits to preprocess the sensing data in parallel to generate sensing data representing environmental conditions, it is possible to avoid all the sensing data being accumulated in the back-end processor for processing, thereby reducing the performance requirements for the processor and reducing the The amount of processing operations performed by the processor.
  • a corresponding memory and processing circuit can be provided in the CS circuit coupled to the interface circuit corresponding to the sensor or adjacent to the CS circuit, so that the sensing data from the sensor can be transmitted to the processing circuit with a short transmission path . This reduces the time spent on transmission and processing and reduces bus network congestion compared to the case where multiple sensor interfaces and multiple processing circuits are coupled to the CS circuit on the ring bus.
  • the sensed data may also be provided directly to
  • the first bus M1 transmits the first sensing data.
  • the camera coupled to the first CS circuit 311 on the first bus M1 transmits captured sensing data to the first processing circuit coupled to the first CS circuit 312 on the first bus M1.
  • the first processing circuit may also be coupled to other CS circuits.
  • the first processing circuit generates first sensory data after processing the image or video data, and transmits to the CS circuit 333 via the first bus M1, and then transmits to the second bus R1 via the bridge circuit B1.
  • other sensing data such as sonar sensing data corresponding to the sonar sensing data from the sonar sensor, can also be transmitted to the second bus R1 via the first bus M1 and the bridge circuit B1.
  • the type of the second bus R1 is different from that of the first bus M1, for example, the second bus R1 is a ring bus.
  • the second bus R1 is a ring bus.
  • Some external electronic devices, such as radar, etc. may be coupled to the CS circuit on the second bus R1 via an interface circuit.
  • the second bus R1 may also be a star or linear bus.
  • each CS site can be coupled to two node circuits.
  • Nodes represent different functional circuits, eg, processors, interface circuits, etc., that can be coupled with the CS circuits on the bus.
  • the second CS circuit 35 on the second bus R1 is coupled to the first node circuit 371 and the second node circuit 372 .
  • the first node circuit 371 is a processor and the second node circuit 372 is an interface circuit.
  • the second bus R1 receives the first sensing data from the first bus M1 via the bridge circuit B1, for example, the image or video sensing data representing the environmental conditions obtained based on the image or video sensing data captured by the camera, and transmits it to the
  • the CS circuit on the second bus R1 is coupled to a node circuit, such as a memory.
  • Another node circuit coupled to the CS circuit on the second bus R1, such as a processor can obtain the sensing data from the memory and process it to generate a control command for controlling the execution device.
  • the sensory data can also be transmitted directly to the processor without going through the memory.
  • sensing data may also be transmitted via the second bus R1.
  • a sorna coupled to a CS circuit on the first bus M1 may transmit sorna sensing data to a processing circuit coupled to the first bus M1, and the processing circuit may process the sensing data to generate sorna sensing data.
  • the sorna sensing data is transmitted to the processor coupled to the second bus R1 via the bridge circuit B1.
  • the radar coupled to the second bus R1 can transmit the sensed data to the second bus R1 via the interface circuit, and the processing circuit coupled to the second bus R1 processes the sensed data Generate radar perception data.
  • the radar sensing data is transmitted to a processor coupled to the second bus R1 via the second bus R1.
  • the present disclosure does not limit the source of the second sensing data transmitted on the second bus R1, which may originate from a sensor coupled to the first bus M1 or from a sensor coupled to the second bus R1.
  • the processor coupled to the second bus R1 may generate control instructions for controlling the executing device based on different types of sensing data after receiving the image or video sensing data and the sorna sensing data. For example, if the image or video sensing data indicates that there is a pedestrian in the center of the environment ahead, and the sorna sensing data also indicates that there is a pedestrian in the center of the environment in front, the processor can generate a braking instruction accordingly, so that the braking device of the vehicle 100 is emergency. Brake to avoid hitting a pedestrian.
  • different types of sensing data for data fusion such as mutual verification, the correctness of the sensing data can be ensured and the safety of vehicles can be improved.
  • the processor can also use different types of other sensory data to comprehensively determine the environmental conditions and generate corresponding control instructions.
  • the processor may use fusion of image or video sensing data and radar sensing data, or fusion of sorna sensing data and radar sensing data to generate control instructions for controlling the execution device. It is also possible to use more kinds of sensing data, such as the above three sensing data, for data fusion to further improve the accuracy of environmental perception, thereby improving decision-making accuracy and safety.
  • a circuit only used for decision-making such as a processor, is configured to be coupled to the ring bus via a CS circuit, thereby reducing resource consumption and transmission delay.
  • some sensors such as radar, may also be coupled to the CS circuit on the second bus R1 via an interface circuit. If there are not many such sensors, placing a small number of sensors on the ring bus will not cause traffic congestion on the ring circuit.
  • the transmission delay can be further reduced and a corresponding fast response can be given.
  • radar sensors are generally used to detect environmental conditions near vehicles (eg, within a few meters or within 1 meter). By placing it on the same loop closer to the processor making the decision, it is possible to respond quickly and avoid vehicle collisions.
  • the first bus M1 and the second bus R1 may work with an asynchronous clock, and the data bit width is different (for example, the data bit width transmitted on the first bus M1 is 1024 bits, while the data bit width transmitted on the second bus R1 The transmitted data bit width is 256 bits), so the bridge circuit B1 is configured to perform asynchronous processing and splice and split data packets.
  • the chip 30 may further include a third bus R2, and the configuration of the third bus R2 is basically the same or similar to that of the second bus R1, so the above aspects for the second bus R1 may be applicable to the second bus R1 Three bus R2.
  • the type of the third bus R2 is the same as that of the second bus R1, for example both are ring buses.
  • the third bus R2 may be a duplicate bus of the second bus R1.
  • the first node circuit 371 and the second node circuit 372 are also coupled to one CS circuit on the third bus R2, and the third bus R2 is coupled to the first bus M1 through the second bridge circuit B2.
  • an auxiliary bus or a duplicate bus such as a third bus R2 is provided.
  • Sensing data sent by the sending device on the first bus M1 can be sent via different paths, wherein the sensing data reaches the first node 371 such as a processor via the bridge circuit B1 and the second bus R1, and the auxiliary data arrives via the bridge circuit B1
  • the circuit B2 and the third bus R2 reach the first node 371 .
  • the received sensing data and assistance data can be verified at the processor to ensure that the received sensing data is correct, as described in further detail below with respect to FIGS. 6 and 7 .
  • FIG. 4 shows a schematic diagram of a CS circuit 35 on a ring bus according to some embodiments of the present disclosure. It can be understood that other CS circuits on the second bus R1 have the same configuration as the CS circuit 35 , so only the CS circuit 35 will be described here as an example.
  • the CS circuit 35 may be coupled with two node circuits, such as a first node circuit 371 and a second node circuit 372, and the CS circuit 35 includes a beat circuit 351, an upload ring buffer (upload ring buffer) SP1, an upload ring buffer SP2 , a download ring buffer (download ring buffer) ME1 and a download ring buffer ME2.
  • the upper ring buffer is configured to buffer data packets to be transmitted, and the lower ring buffer is configured to buffer data packets received from the bus.
  • the tap circuit 351 is configured to route commands and data.
  • FIG. 5 shows a schematic diagram of a mesh bus CS circuit 311 according to some embodiments of the present disclosure. It can be understood that other CS circuits on the first bus M1 have the same configuration as the CS circuit 311, so only the CS circuit 311 will be described here as an example.
  • CS circuit 311 may be coupled to four node circuits N1, N2, N3, and N4, and CS circuit 311 includes four tap circuits P1, P2, P3, and P4, four ring-up buffers SP1, SP2, SP3, and SP4, Four lower ring buffers ME1, ME2, ME3 and ME4, and four ring-exchange buffers TR1, TR2, TR3 and TR4.
  • the beating circuit is configured on the loop bus to upload, send or relay data in a beating manner according to the frequency.
  • the upper ring buffer is here configured to buffer data packets to be transmitted to the horizontal ring bus
  • the lower ring buffer is here to buffer data packets received from the vertical ring bus
  • the ring switching buffer is configured to buffer packets from The data packets transmitted by the horizontal ring bus to the vertical ring bus are buffered.
  • the tap circuit is configured to route commands and data.
  • FIG. 6 shows a schematic diagram of data transmission via a mesh bus according to some embodiments of the present disclosure.
  • Data is transferred from the node circuit N5 to the node circuit N6 via the first bus M1.
  • the node circuit N5 is, for example, an interface circuit coupled with a camera
  • the node circuit N6 is, for example, a memory.
  • the node circuit N5 and the node circuit N6 may also be other devices, such as an AIV circuit, an AIC circuit, and the like.
  • This data is transmitted on the first bus M1 since the node circuit N5 and the node circuit N6 are both coupled to the first bus M1 via the CS circuit.
  • the transmitted data includes control data, payload data, and check data for checking the payload data, such as cyclic redundancy check (cyclic redundancy check, CRC) data.
  • the verification data is used to verify whether the load data is complete and correct.
  • data transmission on the first bus M1 can basically reach the safety level of ASIL-B.
  • the sensing data of different types of ASIL-B levels are further fused in the second bus R1, which can further reach the safety level of ASIL-D.
  • FIG. 7 shows a schematic diagram of data transmission via a ring bus according to some embodiments of the present disclosure.
  • Data is transferred from the node circuit N7 to the node circuit N8 via the second bus R1.
  • node circuit N7 is, for example, a memory
  • node circuit N8 is, for example, a processor.
  • the memory stores at least one of the first sensing data and the second sensing data
  • the memory also stores at least one of the first auxiliary data and the second auxiliary data.
  • the first sensing data is transmitted from the processing circuit on the first bus M1 to the memory on the second bus R1 via the first bus M1 , the bridge circuit B1 and the second bus R1 .
  • First auxiliary data is transferred, for example, from a processing circuit on the first bus M1 to a memory on the third bus R1 via the first bus M1, the bridge circuit B2 and the third bus R2, wherein the memory is located on both the second bus R1 and the third bus on R2.
  • the first auxiliary data is, for example, CRC data for the first sensing data.
  • the first auxiliary data may be the same duplicate sensing data as the first sensing data, or other verification data.
  • Auxiliary data may be used to verify the integrity or correctness of corresponding sensing data to obtain correct environmental data, or combined with sensing data to form correct environmental data.
  • the environmental data is the data describing the internal or surrounding environmental conditions of the vehicle where the chip is located.
  • the environmental conditions include, for example, lane conditions, surrounding buildings, weather conditions, temperature inside the vehicle, and the movement status of nearby pedestrians.
  • the processor may also acquire other data, such as control data, from the memory via the second bus R1, which is not limited in the present disclosure.
  • the second sensing data is transmitted, for example, from the processing circuit on the first bus M1 to the memory on the second bus R1 via the first bus M1 , the bridge circuit B1 and the second bus R1 .
  • the second auxiliary data is for example transferred from the processing circuit on the first bus M1 to the memory on the third bus R2 via the first bus M1, the bridge circuit B2 and the third bus R2.
  • the second sensing data may be transmitted, for example, from another processing circuit on the second bus R1 to a memory on the second bus R1.
  • the second auxiliary data is transferred, for example, from a further processing circuit on the third bus R2 to a memory on the third bus R2, wherein the further processor circuit is located on both the second bus R1 and the third bus R2.
  • the processor may also obtain other data from the memory via the third bus R2, such as a control data copy, which is not limited in the present disclosure.
  • the processor may then determine first environmental data representative of environmental conditions based on the first sensory data and the first assistance data. For example, the processor uses the CRC data to verify that the first sense data is complete and correct. If the first sensing data is complete and correct, the first sensing data can be used as the first environmental data representing the environmental conditions. For example, the first environment data may indicate that there are pedestrians at a certain distance in front of the vehicle, or content of traffic signs ahead. Similarly, the processor may determine second environmental data indicative of environmental conditions based on the second sensory data and the second assistance data. The transmission paths of the second sensing data and the second auxiliary data may be similar to the transmission paths of the first sensing data and the first auxiliary data respectively, which will not be repeated here.
  • the types of the second sensing data and the first sensing data are different.
  • the first sensing data may be image or video data captured by a camera
  • the second sensing data may be data derived from LiDAR measurements.
  • the processor can then perform data fusion of different types of sensory data and determine environmental conditions. For example, when both the first sensing data and the second sensing data indicate that there is a pedestrian in the middle of the road ahead, the processor may generate fusion data based on the first sensing data and the second sensing data and generate a control for controlling the execution device based on the fusion data. Signal, e.g. to control the braking device for emergency braking.
  • the processor can also make the vehicle display screen display pedestrians on the road ahead based on the fusion data.
  • the processor may directly generate a control signal for controlling the actuator based on the first sensing data and the second sensing data.
  • the processor After the processor generates the control signal, it also transmits the control signal to the interface circuit coupled with the execution device through the second bus R1 and the third bus R2 respectively, and the interface circuit is also coupled to the second bus R1 and the third bus R2 as a node circuit.
  • the processor may also transmit control signals to the execution device through a dedicated connection line.
  • Fig. 8 shows a schematic flowchart of a method 800 for generating a control signal according to some embodiments of the present disclosure.
  • the method 800 can be implemented by the chip 30 described above. Therefore, various aspects described above with respect to the chip 30 can be applied to the method 800 , which will not be repeated here.
  • first sensing data is transmitted through a first bus.
  • the first bus is, for example, a mesh bus, and the first sensing data is, for example, camera data.
  • the interface circuit coupled with the first bus receives original image or video data from the camera, and the original image or video data is transmitted to a processing circuit on the first bus, such as AIV or AIC, for example via the first bus.
  • the processing circuit obtains the first perception data after processing the original image or video data.
  • the first sensing data is transmitted to the second bus via the first bus and the bridge circuit.
  • the type of the second bus is different from the type of the first bus
  • the type of the second sensory data is different from the type of the first sensory data different.
  • the second bus is, for example, a ring bus
  • the second sensing data is, for example, LiDAR data.
  • the LiDAR can be coupled to the first bus or the second bus via an interface circuit. In the case of being coupled to the first bus, the LiDAR data is processed and transmitted in a manner similar to that of the camera data, which will not be repeated here.
  • the LiDAR transmits the raw LiDAR data via the second bus to a processing circuit, such as an AIV or AIC, to generate second sensing data.
  • the second sensing data is transmitted to the processor on the second bus through the second bus.
  • a control signal for controlling an execution device is generated based on the first sensing data and the second sensing data acquired from the second bus.
  • the processor may, for example, generate the control signal based on the first sensing data and the second sensing data in the previously described manner. For example, the processor may perform data fusion of the first sensing data and the second sensing data to generate a control signal.
  • the processor may further determine the first environmental data based on the first sensing data and the first auxiliary data, and determine the second environmental data based on the second sensing data and the second auxiliary data, and then combine the first environmental data and the The second environmental data is fused to determine environmental conditions, and corresponding control signals are generated based on the determined environmental conditions.
  • the processor may directly determine the control signal based on the first environment data and the second environment data.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Traffic Control Systems (AREA)

Abstract

本公开涉及一种芯片、电子组件、交通工具和用于生成控制信号的方法。芯片包括网状总线和环形总线。网状总线耦合至各个传感器以接收和传输感知数据。环形总线耦合至处理器和网状总线。处理器接收诸如来自摄像头和激光雷达之类的不同传感器的不同类型的感知数据，并且进行数据融合以生成用于控制执行装置的控制信号。在本公开中，通过在单个芯片内将数量众多的感知传感器耦合至网状总线并且将数量较少的控制器耦合至环形总线，可以分别利用各个总线的优势，降低成本和资源消耗。此外，通过将不同类型的数据进行融合，可以降低硬件成本需求并且同时提升芯片和使用该芯片的交通工具的安全等级。

Description

用于交通工具的集成电路芯片 技术领域
本公开涉及电子领域，更具体而言涉及用于交通工具的集成电路芯片和电子组件。
背景技术
随着信息技术的发展，诸如智能汽车之类的智能交通工具越来越多。对于智能交通工具而言，诸如自动驾驶、语音交互和驾驶员检测之类的功能成为智能交通工具的核心竞争力。这些功能的实现往往依赖于相应硬件，例如车载芯片、摄像头、光学雷达(light detection and ranging,LiDAR)、声纳(sorna)、雷达(radar)、麦克风等。
另一方面，对于自动驾驶而言，其与人身安全直接相关，因此用于实现自动驾驶的芯片的安全性尤为关注。一些常规自动驾驶方案是通过双芯片的冗余设计来提升安全规格。然而，双芯片的设计方案往往成本较高并且资源消耗较大。
发明内容
鉴于上述问题，本公开的实施例旨在提供一种芯片、电子组件、交通工具和用于生成控制信号的方法，用于以低成本的方式实现交通工具的控制。
根据本公开的第一方面，提供一种芯片。芯片包括第一总线、第二总线和处理器。第一总线被配置为传输第一感知数据。第二总线耦合至第一总线。第二总线的类型与第一总线的类型不同。第二总线被配置为传输第二感知数据和来自第一总线的第一感知数据。第二感知数据的类型与第一感知数据的类型不同。处理器耦合至第二总线，并且处理器被配置为基于第一感知数据和第二感知数据生成用于控制执行装置的控制信号。通过将各个功能电路基于功能类型(例如感测和决策)分别不同总线耦合，可以在第一总线上传输感知类数据，并且在第二总线上传输决策数据，从而可以分别有效利用各个总线的优势，例如环形总线的低延迟和高效数据传输以及网格总线的扩展性强、大带宽和低延迟，并且减少资源消耗和延迟。
在一种实现方式中，第一总线选自包括网状总线、全连接总线和星形总线的组中。第二总线选自包括环形总线、线形总线和树形总线的组中。
在一种实现方式中，芯片还包括第一处理电路。第一处理电路耦合至第一总线，并且接收来自耦合至第一总线的第一传感器的第一感测数据。第一处理电路对第一感测数据进行处理以生成第一感知数据。通过对感测数据进行本地处理，可以避免传输具有大的数据量的原始感测数据，从而节省带宽和成本。
在一种实现方式中，第一处理电路选自包括人工智能矢量电路和人工智能核芯电路的组。第一传感器选自包括摄像头、光学雷达、声纳和雷达的组中。
在一种实现方式中，芯片还包括第三总线。第三总线耦合至第一总线。第三总线的类型与第二总线相同。第三总线被配置为传输来自第一总线的与第一感知数据对应的第一辅助数据和与第二感知数据对应的第二辅助数据。处理器耦合至第三总线并且处理器被进一步配置为基于第一感知数据、第一辅助数据、第二感知数据和第二辅助数据生成控制信号。通过使用与第二总线相同类型的第三总线来传输辅助数据，可以用于验证在第二总线上传输的感知数据的完整性或正确性以提高数据传输的安全性，从而提升芯片的安全性能。
在一种实现方式中，辅助数据包括循环冗余校验数据。在一种实现方式中，芯片还包括第二处理电路。第二处理电路耦合至第一总线或第二总线，并且接收来自耦合至第一总线或第二总线的第二传感器的第二感测数据。第二处理电路对第二感测数据进行处理以生成第二感知数据。通过对感测数据进行本地处理，可以避免传输具有大的数据量的原始感测数据，从而节省带宽和成本。
在一种实现方式中，第二处理电路选自包括人工智能矢量电路和人工智能核芯电路的组。第二传感器选自包括摄像头、光学雷达、声纳和雷达的组中。
在一种实现方式中，处理器被进一步配置为基于第一感知数据和第一辅助数据确定用于表示环境状况的第一环境数据；基于第二感知数据和第二辅助数据确定用于表示环境状况的第二环境数据；以及基于第一环境数据和第二环境数据生成控制信号。通过使用不同类型的环境数据，可以交互验证环境数据彼此的正确性，例如如果第一环境数据表示前方道路有行人，并且不同类型的第二环境数据也表示前方道路有行人，则可以基于一致性的结果生成对应的控制信号。
在一种实现方式中，处理器被进一步配置为将第一环境数据和第二环境数据进行融合以生成融合数据；以及基于融合数据生成控制信号。在一种实现方式中，处理器被进一步配置为将第一感知数据和第二感知数据进行融合以生成融合数据；以及基于融合数据生成控制信号。通过将诸如摄像头数据和LiDAR数据之类的不同类型的感知数据进行融合，可以验证感知数据的正确性，确保感测到的正确的环境对象并且生成对应的决策或控制信号，以使得交通工具的诸如致动器之类的执行装置可以做出相应的正确操作。这样，可以提高交通工具的安全等级。
在一种实现方式中，芯片还包括：第一交叉站点电路，位于第一总线上并且被配置为传输数据；以及第一处理电路，耦合至第一交叉站点电路并且被配置为基于来自第一传感器的第一感测数据生成第一感知数据，第一传感器耦合至第一交叉站点电路。通过在同一交叉站点电路耦合传感器和处理电路，可以避免大量感测数据在第一总线上的长距离传输，从而减少带宽拥堵并且提高处理速度。
在一种实现方式中，芯片还包括第二交叉站点电路，位于第二总线上并且被配置为传输数据；以及第二处理电路，耦合至第二交叉站点电路并且被配置为基于来自第二传感器的第二感测数据生成第二感知数据，第二传感器耦合至第二交叉站点电路。通过在同一交叉站点电路耦合传感器和处理电路，可以避免大量感测数据在第一总线上的长距离传输，从而减少带宽拥堵并且提高处理速度。
在一种实现方式中，第一总线包括网状总线，以及第二总线包括环形总线，以及第三总线包括环形总线。
根据本公开的第二方面，提供一种电子组件。电子组件包括电路板以及根据第一方面的芯片。芯片被安装在电路板上。通过将芯片中的各个功能电路基于功能类型(例如感测和决策)分别不同总线耦合，可以在第一总线上传输感知类数据，并且在第二总线上传输决策数据，从而可以分别有效利用各个总线的优势，例如环形总线的低延迟和高效数据传输以及网格总线的扩展性强、大带宽和低延迟，并且减少资源消耗和延迟。
根据本公开的第三方面，提供一种交通工具。交通工具包括电源；以及根据第二方面的电子组件。电子组件由电源供电。将交通工具中的芯片中的各个功能电路基于功能类型(例如感测和决策)分别不同总线耦合，可以在第一总线上传输感知类数据，并且在第二总线上 传输决策数据，从而可以分别有效利用各个总线的优势，例如环形总线的低延迟和高效数据传输以及网格总线的扩展性强、大带宽和低延迟，并且减少资源消耗和延迟。
在一种实现方式中，交通工具还包括第一传感器和第二传感器。第一传感器经由第一接口耦合至第一总线并且被配置为生成第一传感器数据并且将第一传感器数据传输至第一总线。第二传感器经由第二接口耦合至第二总线，并且配置为生成第二传感器数据并且将第二传感器数据传输至第二总线。
根据本公开的第四方面，提供一种用于生成控制信号的方法。该方法包括通过第一总线传输第一感知数据；通过第二总线传输第二感知数据和来自于第一总线的第一感知数据，第二总线的类型与第一总线的类型不同，第二感知数据的类型与第一感知数据的类型不同；基于从第二总线获取的第一感知数据和第二感知数据生成用于控制执行装置的控制信号。将芯片中的各个功能电路基于功能类型(例如感测和决策)分别不同总线耦合，可以在第一总线上传输感知类数据，并且在第二总线上传输决策数据，从而可以分别有效利用各个总线的优势，例如环形总线的低延迟和高效数据传输以及网格总线的扩展性强、大带宽和低延迟，并且减少资源消耗和延迟。
在一种实现方式中，方法还包括通过第三总线传输来自第一总线的与第一感知数据对应的第一辅助数据和与第二感知数据对应的第二辅助数据，第三总线耦合至第一总线，第三总线的类型与第二总线相同；以及基于第一感知数据、第一辅助数据、第二感知数据和第二辅助数据生成控制信号。通过使用与第二总线相同类型的第三总线来传输辅助数据，可以用于验证在第二总线上传输的感知数据的完整性或正确性以提高数据传输的安全性，从而提升芯片的安全性能。
在一种实现方式中，基于从第二总线获取的第一感知数据和第二感知数据生成用于控制执行装置的控制信号包括基于第一感知数据和第一辅助数据确定用于表示环境状况的第一环境数据；基于第二感知数据和第二辅助数据确定用于表示环境状况的第二环境数据；以及基于第一环境数据和第二环境数据生成控制信号。通过使用不同类型的环境数据，可以交互验证环境数据彼此的正确性，例如如果第一环境数据表示前方道路有行人，并且不同类型的第二环境数据也表示前方道路有行人，则可以基于一致性的结果生成对应的控制信号。
在一种实现方式中，基于第一环境数据和第二环境数据生成控制信号包括将第一环境数据和第二环境数据进行融合以生成融合数据；以及基于融合数据生成控制信号。。通过使用不同类型的环境数据，可以交互验证环境数据彼此的正确性，例如如果第一环境数据表示前方道路有行人，并且不同类型的第二环境数据也表示前方道路有行人，则可以基于一致性的结果生成对应的控制信号。
应当理解，发明内容部分中所描述的内容并非旨在限定本公开的实施例的关键或重要特征，亦非用于限制本公开的范围。本公开的其它特征将通过以下的描述变得容易理解。
附图说明
结合附图并参考以下详细说明，本公开各实施例的上述和其他特征、优点及方面将变得更加明显。在附图中，相同或相似的附图标记表示相同或相似的元素，其中：
图1示出了根据本公开的一些实施例的环境示意图；
图2示出了根据本公开的一些实施例的电子组件的示意框图；
图3示出了根据本公开的一些实施例的芯片的示意图；
图4示出了根据本公开的一些实施例的环形总线交叉站点的示意图；
图5示出了根据本公开的一些实施例的网状总线交叉站点的示意图；
图6示出了根据本公开的一些实施例的经由网状总线的数据传输的示意图；
图7示出了根据本公开的一些实施例的经由环形总线的数据传输的示意图；以及
图8示出了根据本公开的一些实施例的用于生成控制信号的方法的示意流程图。
具体实施方式
下面将参照附图更详细地描述本公开的实施例。虽然附图中显示了本公开的某些实施例，然而应当理解的是，本公开可以通过各种形式来实现，而且不应该被解释为限于这里阐述的实施例，相反提供这些实施例是为了更加透彻和完整地理解本公开。应当理解的是，本公开的附图及实施例仅用于示例性作用，并非用于限制本公开的保护范围。
在本公开的实施例的描述中，术语“包括”及其类似用语应当理解为开放性包含，即“包括但不限于”。术语“基于”应当理解为“至少部分地基于”。术语“一个实施例”或“该实施例”应当理解为“至少一个实施例”。术语“第一”、“第二”等等可以指代不同的或相同的对象。术语“和/或”表示由其关联的两项的至少一项。例如“A和/或B”表示A、B、或者A和B。下文还可能包括其他明确的和隐含的定义。
应理解，本申请实施例提供的技术方案，在以下具体实施例的介绍中，某些重复之处可能不再赘述，但应视为这些具体实施例之间已有相互引用，可以相互结合。
用于交通工具的芯片的安全性能尤其受到重视。ISO 26262标准对功能安全划分为A级、B级、C级、D级这4个安全等级(Automotive Safety Integrity Level，ASIL)，其中ASIL-D为最高等级，代表最严苛的安全要求。在一些常规方案中，通过使用双芯片的冗余设计来提升安全性。每个芯片通常集成了众多功能电路，例如处理器、存储器、缓冲器、人工智能矢量(artificial intelligence vector,AIV)电路、人工智能核芯(artificial intelligence core,AIC)电路、接口电路等，其中每种功能电路的数量可能为一个或多个。这些功能电路分别耦合至位于总线上的多个交叉站点(cross statation,CS)电路，并且多个CS电路可以通过总线进行互连。由于总线上的各个电路的频率并不完全相同，因此总线上的CS电路按照频率以打拍的方式对数据进行接收和发送。例如，CS电路可以从总线上接收经由总线传输的数据并且以打拍的方式将该数据传输至与其相连的功能电路，或以打拍的方式将来自功能电路的数据传输至总线上的其它功能电路。在CS电路并不与源电路和目的地电路直接耦合的情形下，该CS电路可以以打拍的方式在总线上对数据进行中继传输。CS电路使得各个功能电路在彼此之间可以经由总线传递指令和数据。对于使用双芯片冗余设计的架构而言，这要求每个芯片都具有基本相同的配置以实现冗余的相同功能。此外，每个芯片内部通常还具有相同类型的主总线和副总线，例如网状总线，以确保在每个芯片内部的各个电路彼此之间的数据的安全性，例如确保数据在芯片内部传输不会丢失或产生错误。因此，在一些常规方案中，实际上具有相同类型的四总线的双芯片架构以确保安全性。
一种常规总线类型为环形(ring)总线，环形总线上的每个CS电路可以耦合2个功能电路。在常规的环形总线上的功能电路相对较少时，环形总线具有较低的延迟和快速高效的数据传输。然而当芯片***中的功能电路增多并且导致CS电路增加时，各个功能电路之间的传输延迟变大。另一方面，由于环形总线的承载带宽有限，因此当各个功能电路之间有大量数据传输需求时，会导致环形电路的业务拥堵并且带宽受限。
另一种常规总线类型为网状(mesh)总线，网状总线上的每个CS电路可以耦合4个功能电路。网状总线由多条横向环形总线和多条纵向环形总线彼此交叉形成的二维总线网络。网状总线可扩展性强，并且网状总线上的功能电路彼此之间可以有大带宽和低延迟。然而，当所有的功能电路均经由CS电路耦合至网状总线时，网状总线的结构过于复杂，资源消耗大并且延迟也较大。此外，由于在常规设计中往往采用双芯片的冗余设计，因此这会导致成本和资源消耗更大。
在本公开的一些实施例中，通过将各个功能电路基于功能类型(例如感测和决策)分别与单个芯片内的诸如网状总线之类的第一总线和诸如环形总线之类的第二总线耦合，可以在第一总线上传输感知类数据，并且在第二总线上传输决策数据，从而可以分别有效利用各个总线的优势，例如环形总线的低延迟和高效数据传输以及网格总线的扩展性强、大带宽和低延迟，并且减少资源消耗和延迟。此外，通过将诸如摄像头数据和LiDAR数据之类的不同类型的感知数据进行融合，可以验证感知数据的正确性，确保感测到的正确的环境对象并且生成对应的决策或控制信号，以使得交通工具的诸如致动器之类的执行装置可以做出相应的正确操作。这样，可以提高交通工具的安全等级，例如达到ASIL-D级。
图1示出了根据本公开的一些实施例的环境示意图。在该环境示意图中，交通工具100包括电子组件20。在一个实施例中，交通工具100例如是汽车。备选地，交通工具100也可以是其它交通工具，例如摩托车、空中交通工具或是水面交通工具等。交通工具100还包括其它部件，例如各种传感器和执行装置。各种传感器和执行装置可以电耦合至电子组件20。传感器可以例如包括摄像头、LiDAR、sorna、radar、温度传感器、湿度传感器等，以感测环境数据，并且将所感测到的数据传输至电子组件20中的芯片。电子组件20中的芯片基于所接收的各种数据进行分析、计算(compute)和处理并且生成决策信号或控制信号，例如交通工具启动信号、制动信号、转向信号和开启空调信号等。诸如致动器、刹车和车载空调之类的执行装置可以接收来自电子组件20中的芯片的决策信号或控制信号以分别执行各种操作，例如启动、刹车、转向、开启空调等。
图2示出了根据本公开的一些实施例的电子组件20的示意框图。电子组件20包括电路板21以及安装在电路板上的多个电子部件。电路板21包括印刷电路板(printed circuit board,PCB)和柔性电路板(flexbile printed circuit,FPC)中的至少一个。多个电子部件包括通信芯片22、匹配电路24、电源管理芯片26和芯片30，其中通信芯片22用于与外部设备或其是其它电子装置进行通信，匹配电路24用于对各个电子部件进行匹配，电源管理芯片26例如可以耦合到电源以对所接收的电力进行管理并且将电力分配给各个电子部件。芯片30用于接收上述传感器的感测数据以生成感知数据并且基于感知数据生成控制信号。感测数据表示由传感器所检测到的表示环境状况的原生数据，而感知数据则表示对感测数据进行处理之后表征或描述环境状况的经处理的数据。例如，摄像头所拍摄的原生图像或视频数据是感测数据，但是对图像或视频进行处理并且通过AI算法识别并判断出所拍摄的图像或视频中存在行人并且对此进行表征或描述的数据为感知数据。虽然在图2中示出了四个电子部件，但是可以理解电子组件20可以包括更多或更少的电子部件，本公开对此不进行限制。
图3示出了根据本公开的一些实施例的芯片30的示意图。在一个实施例中，芯片30包括第一总线M1、第二总线R1和第三总线R2。芯片30还包括在第一总线M1上的多个第一CS电路，在第二总线R1上的多个第二CS电路以及在第三总线R2上的多个第三CS电路。芯片30还包括耦合在第一总线M1和第二总线R1之间的桥接电路B1，以及耦合在第一总线 M1和第三总线R2之间的桥接电路B2。芯片30还包括第一节点电路371、第二节点电路372和其它未示出的电路。虽然在图3中示出了芯片30的一种具体配置方式，但是可以理解芯片30可以包括更多或更少的电路，本公开对此不进行限制。此外，虽然使用桥接电路将不同总线进行耦合，但是本公开对此不做限制。用于耦合不同总线的其它方式是可能的。例如，诸如存储器之类一些功能电路充当耦合电路。第一总线M1上的感知数据被存储至存储器，并且第二总线R1上的处理器从存储器获取感知数据。
在一个实施例中，第一总线M1例如是网状总线。在图3中，第一总线M1例如由3个横向环形总线和4个纵向环形总线彼此交叉形成，并且每个交叉点处设置有一个CS电路。例如，第一横向环形总线和4个纵向环形总线的交叉点处设置有CS电路311、312、313和314，并且第一纵向环形总线和3个横向环形总线的交叉点处设置有CS电路311、321和331。虽然在图3中以3×4示出了一种具体的网状总线的配置，但是这仅是示意而非对本公开的范围进行限制。横向总线和纵向总线可以具有更多或更少的数目，并且横向总线的数目和纵向总线的数目可以相同或不同。备选地，第一总线M1也可以是其它类型的总线，例如全连接总线。
第一总线M1上的每个CS电路基本上相同，因此在下文仅以单个CS电路进行描述。可以理解，对该单个CS电路的描述可以应用至第一总线M1上的其它CS电路。在第一总线M1为网状总线的情形下，每个CS电路可以耦合至四个节点(node)电路。节点电路表示可以与总线上的CS电路耦合的不同功能电路，例如，AIC电路、AIV电路、接口电路、缓冲器、等等。本公开不对节点电路进行任何限制。一些外部电子装置，例如LiDAR、摄像头、sorna等，可以经由接口电路耦合至CS电路。在网状电路的情形下，由于各个节点之间具有多条通信线路，因此在操作时可以根据需要选择可用的最近通信路径以在网状总线上传输数据，从而节省传输时间并且提供效率。虽然通过使用CS电路将各个功能电路耦合在总线上，但是本公开对此不做限制。使用其它方式将各个功能电路与总线耦合的方式是可能的。
交通工具100通常具有多个环境状况传感器以从多个维度感测交通工具四周的环境状况。例如，交通工具100可以具有12个摄像头、4个LiDAR、12个radar和4个sonar。在一个实施例中，与上述多个传感器对应的各个接口电路可以耦合至网状电路上的各个CS电路。芯片30可以具有分别与位于第一总线M1上的多个CS电路耦合的多个诸如AIV和/或AIC之类的处理电路。与环形总线相比，这可以避免环形总线在多个接入电路时容易出现的拥堵和带宽有限的情形。
实时感测的数据可以被存储在存储器中，然后由诸如AIV和/或AIC之类的处理电路从存储器读取数据进行初步处理，以生成表征环境状况的感知数据。通过由多个处理电路对感测数据进行并行预处理以生成表征环境状况的感知数据，可以避免所有感测数据都堆积到后端处理器处理，从而可以减小对于处理器的性能要求并且降低处理器的处理操作量。在一个实施例中，可以在与传感器对应的接口电路所耦合的CS电路或邻近CS电路设置对应的存储器和处理电路，从而使得来自传感器的感测数据可以以较短的传输路径传递至处理电路。相比于多个传感器接口和多个处理电路耦合到环形总线上的CS电路的情形，这减少传输和处理所消耗的时间，并且减少总线网络拥堵。备选地，感测数据也可以直接提供给处理电路。
在图3的实施例中，第一总线M1传输第一感知数据。例如，与第一总线M1上的第一CS电路311耦合的摄像头将所拍摄的感测数据传输至与第一总线M1上的第一CS电路312耦合的第一处理电路。备选地，第一处理电路也可以耦合至其它CS电路。第一处理电路对 图像或视频数据进行处理之后生成第一感知数据，并且经由第一总线M1传输至CS电路333，并且随后经由桥接电路B1传输至第二总线R1。此外，其它的感知数据，例如与来自声纳传感器的声纳感测数据对应的声纳感知数据，也可以经由第一总线M1、桥接电路B1传输至第二总线R1。
在一个实施例中，第二总线R1的类型与第一总线M1的类型不同，例如第二总线R1是环形总线。本公开不对节点进行任何限制。一些外部电子装置，例如radar等，可以经由接口电路耦合至第二总线R1上的CS电路。备选地，第二总线R1也可以是星形或线性总线。
第二总线R1上具有多个CS站点，每个CS站点可以耦合两个节点电路。节点表示可以与总线上的CS电路耦合的不同功能电路，例如，处理器、接口电路、等等。如图3所示，第二总线R1上的第二CS电路35与第一节点电路371和第二节点电路372耦合。在一个实施例中，第一节点电路371是处理器，第二节点电路372是接口电路。在一个实施例中，第二总线R1上可以有比图3所示出的5个CS电路更多或更少的CS电路。第二总线R1经由桥接电路B1接收来自第一总线M1的第一感知数据，例如基于由摄像头拍摄的图像或视频感测数据得到的表征环境状况的图像或视频感知数据，并且将其传输至与第二总线R1上的CS电路耦合的节点电路，例如存储器。与第二总线R1上的CS电路耦合的另一节点电路，例如处理器，可以从存储器获取感知数据并且进行处理以生成用于控制执行装置的控制指令。备选地，感知数据也可以被直接传输至处理器，而不经由存储器。
此外，其它类型的感知数据也可以经由第二总线R1进行传输。例如，与第一总线M1上的一个CS电路耦合的sorna可以将sorna感测数据传输至与第一总线M1耦合的处理电路，并且处理电路将该感测数据处理之后生成sorna感知数据。sorna感知数据经由桥接电路B1传输至与第二总线R1耦合的处理器。在另一个实施例中，与第二总线R1耦合的radar可以将所感测的感测数据经由接口电路传输至第二总线R1上，并且与第二总线R1耦合的处理电路在处理感测数据之后生成radar感知数据。radar感知数据经由第二总线R1传输至与第二总线R1耦合的处理器。本公开不对在第二总线R1上传输的第二感知数据的来源进行限制，其可以源于与第一总线M1耦合的传感器或是源于与第二总线R1耦合的传感器。
与第二总线R1耦合的处理器在接收到图像或视频感知数据和sorna感知数据之后可以基于不同类型的感知数据来生成用于控制执行装置的控制指令。例如，如果图像或视频感知数据表示前方环境正中央存在行人，并且sorna感知数据也表示前方环境正中央存在行人，则处理器可以相应地生成制动指令，以使得交通工具100的制动装置紧急制动以避免碰撞行人。通过使用不同类型的感知数据进行数据融合，例如互相验证，可以确保感知数据的正确性并且提高交通工具的安全性。此外，由于使用不同类型的感知数据进行数据融合，因此无需将硬件装置进行冗余配置(例如成倍设置摄像头、sorna或radar)，从而减少了硬件配套成本。备选地，处理器也可以使用不同类型的其它感知数据来综合确定环境状况并且生成相应控制指令。例如，处理器可以使用图像或视频感知数据和radar感知数据的融合、或sorna感知数据和radar感知数据的融合来生成用于控制执行装置的控制指令。也可以使用更多种感知数据，例如上述三种感知数据，进行数据融合以进一步提高环境感知的准确性，从而提高决策正确性和安全性。
在图3的实施例中，仅用于决策的电路，例如处理器，被设置为经由CS电路耦合至环形总线，由此可以减小资源消耗和传输延迟。备选地，基于总线业务负荷或性能优化需要，也可以有一些传感器，例如radar，经由接口电路耦合至第二总线R1上的CS电路。如果这 类传感器数量不多，则将少量传感器布置在环形总线上并不会引起环形电路的流量拥堵。此外，由于其被布置在邻近处理器的节点，可以进一步降低传输延迟并且给出相应的快速响应。例如，radar传感器通常用于检测交通工具附近(例如几米或1米以内)的环境状况。通过将其设置在更为靠近做出决策的处理器的相同环路上，可以快速做出响应，避免交通工具碰撞的产生。
在一些实施例中，由于第一总线M1和第二总线R1可能以异步时钟工作，并且数据位宽不同(例如，第一总线M1上传输的数据位宽是1024比特，而第二总线R1上传输的数据位宽是256比特)，因此桥接电路B1被配置为进行异步处理以及对数据包进行拼接和拆分处理。
在另一些实施例中，芯片30还可以包括第三总线R2，第三总线R2的配置与第二总线R1的配置基本上相同或相似，因此上面针对第二总线R1的各个方面可以适用于第三总线R2。例如，第三总线R2的类型与第二总线R1相同，例如两者均为环形总线。第三总线R2可以是第二总线R1的副本总线。如图3所示，第一节点电路371和第二节点电路372也耦合至第三总线R2上的一个CS电路，并且第三总线R2通过第二桥接电路B2耦合至第一总线M1。通过设置副本总线，可以进一步提高芯片30的安全性，例如满足ASIL-D的要求。
感知数据从第一总线M1上的发送装置(例如，AIV电路或AIC电路)到第二总线R1上的接收装置(例如处理器)的传输过程中可能会被丢失或产生错误。通过不同类型的数据融合，可以在一定程度上克服上述问题，使得处理器仍能做出正确决策。为了进一步提供交通工具的安全性，在一个实施例中，设置辅助总线或副本总线，例如第三总线R2。由第一总线M1上的发送装置所发送的感知数据可以经由不同路径发送，其中感知数据经由桥接电路B1和第二总线R1到达诸如处理器之类的第一节点371，而辅助数据则经由桥接电路B2和第三总线R2到达第一节点371。通过使用两个不同的传输路径，可以在处理器处对所接收的感知数据和辅助数据进行验证，以确保所接收的感知数据是正确的，如下文针对图6和图7进一步具体描述。
图4示出了根据本公开的一些实施例的环形总线上的CS电路35的示意图。可以理解，第二总线R1上的其它CS电路具有与CS电路35相同的配置，因此在此仅以CS电路35为例进行描述。CS电路35可以与两个节点电路耦合，例如第一节点电路371和第二节点电路372，并且CS电路35包括打拍电路351、上环缓冲器(upload ring buffer)SP1、上环缓冲器SP2、下环缓冲器(download ring buffer)ME1和下环缓冲器ME2。上环缓冲器被配置为将待传输的数据包进行缓存，而下环缓冲器被配置为对从总线接收的数据包进行缓存。打拍电路351则被配置为对命令和数据进行路由传输。
图5示出了根据本公开的一些实施例的网状总线CS电路311的示意图。可以理解，第一总线M1上的其它CS电路具有与CS电路311相同的配置，因此在此仅以CS电路311为例进行描述。CS电路311可以耦合至四个节点电路N1、N2、N3和N4，并且CS电路311包括四个打拍电路P1、P2、P3和P4、四个上环缓冲器SP1、SP2、SP3和SP4、四个下环缓冲器ME1、ME2、ME3和ME4、以及四个换环缓冲器(ring-exchange buffer)TR1、TR2、TR3和TR4。打拍电路被配置在环路总线上按照频率以打拍的方式对数据进行上传、下发或中继传输。上环缓冲器在此被配置为对将传输至横向环状总线的数据包进行缓存，下环缓冲器在此对从纵向环状总线接收的数据包进行缓存，换环缓冲器被配置为从横向环状总线向纵向环状总线传输的数据包进行缓冲。打拍电路则被配置为对命令和数据进行路由传输。
图6示出了根据本公开的一些实施例的经由网状总线的数据传输的示意图。数据从节点 电路N5经由第一总线M1向节点电路N6进行传输。在一个实施例中，节点电路N5例如是与摄像头耦合的接口电路，并且节点电路N6例如是存储器。节点电路N5和节点电路N6也可以是其它器件，例如AIV电路、AIC电路等。由于节点电路N5和节点电路N6均经由CS电路耦合至第一总线M1，因此该数据在第一总线M1上传输。在一个实施例中，所传输的数据包括控制数据、载荷数据和用于校验载荷数据的校验数据，例如循环冗余校验(cyclic redundancy check，CRC)数据。校验数据用于校验载荷数据是否完整和正确。通过使用数据校验，数据在第一总线M1上的传输基本上可以达到ASIL-B的安全等级。不同类型的ASIL-B等级的感知数据在第二总线R1进一步融合，可以进一步达到ASIL-D的安全等级。
图7示出了根据本公开的一些实施例的经由环形总线的数据传输的示意图。数据从节点电路N7经由第二总线R1向节点电路N8进行传输。在一个实施例中，节点电路N7例如是存储器，并且节点电路N8例如是处理器。存储器存储了第一感知数据和第二感知数据中的至少一种，并且存储器还存储了第一辅助数据和第二辅助数据中的至少一种。第一感知数据例如从第一总线M1上的处理电路经由第一总线M1、桥接电路B1和第二总线R1传输至第二总线R1上的存储器。第一辅助数据例如从第一总线M1上的处理电路经由第一总线M1、桥接电路B2和第三总线R2传输至第三总线R1上的存储器，其中存储器同时位于第二总线R1和第三总线R2上。第一辅助数据例如是针对第一感知数据的CRC数据。备选地，第一辅助数据可以是与第一感知数据相同的副本感知数据，或是其它校验数据。辅助数据可以用于验证对应的感知数据的完整性或正确性以获得正确的环境数据，或与感知数据组合以形成正确的环境数据。环境数据是描述芯片所在的交通工具内部或周围环境状况的数据，环境状况例如包括车道状况、周围的建筑物、天气状况、车内温度、附近行人的运动状态等。处理器还可以经由第二总线R1从存储器获取其它数据，例如控制数据，本公开对此不进行限制。
类似地，第二感知数据例如从第一总线M1上的处理电路经由第一总线M1、桥接电路B1和第二总线R1传输至第二总线R1上的存储器。第二辅助数据例如从第一总线M1上的处理电路经由第一总线M1、桥接电路B2和第三总线R2传输至第三总线R2上的存储器。备选地，第二感知数据例如可以从第二总线R1上的另一处理电路传输至第二总线R1上的存储器。第二辅助数据例如从第三总线R2上的另一处理电路传输至第三总线R2上的存储器，其中该另一处理器电路同时位于第二总线R1和第三总线R2上。处理器还可以经由第三总线R2从存储器获取其它数据，例如控制数据副本，本公开对此不进行限制。
处理器继而可以基于第一感知数据和第一辅助数据确定表示环境状况的第一环境数据。例如，处理器使用CRC数据验证第一感知数据是否完整和正确。如果第一感知数据完整和正确，则可以将第一感知数据用作表示环境状况的第一环境数据。例如，第一环境数据可以表示交通工具前方一定距离处存在行人、或者前方交通标志的内容。类似地，处理器可以基于第二感知数据和第二辅助数据确定表示环境状况的第二环境数据。第二感知数据和第二辅助数据的传输路径可以分别与第一感知数据和第一辅助数据的传输路径相似，在此不再赘述。
第二感知数据和第一感知数据的类型不同。例如，第一感知数据可以是源自于摄像头拍摄的图像或视频数据，第二感知数据可以是源自LiDAR测量的数据。处理器继而可以对不同类型的感知数据进行数据融合，并且确定环境状况。例如，当第一感知数据和第二感知数据均表示前方道路中央存在行人时，处理器可以基于第一感知数据和第二感知数据生成融合数据并且基于该融合数据生成用于控制执行装置的控制信号，例如控制制动装置进行紧急制动。此外，处理器还可以基于该融合数据使得车载显示屏显示在前方道路中的行人。备选地，处 理器可以基于第一感知数据和第二感知数据直接生成用于控制执行装置的控制信号。
处理器在生成控制信号之后也分别通过第二总线R1和第三总线R2传输控制信号至与执行装置耦合的接口电路，该接口电路也作为节点电路耦合至第二总线R1和第三总线R2。备选地，处理器也可以通过专用连接线路传输控制信号至执行装置。通过将用于决策的数据分别经由第二总线R1和第三总线R2传输至处理器，处理器可以确保所接收到的感知数据的完整性和正确性，从而进一步提高交通工具的智能芯片所做的决策的准确率并且相应地提高交通工具的安全性能。例如，通过不同类型的数据融合以及通过不同的第二总线和第三总线传输感知数据和控制信号，可以使得芯片30整体达到ASIL-D的安全级别。
图8示出了根据本公开的一些实施例的用于生成控制信号的方法800的示意流程图。方法800可以由前面所描述的芯片30实现。因此，上文针对芯片30所描述的各个方面可以应用于方法800，在此不再赘述。在802，通过第一总线传输第一感知数据。第一总线例如为网状总线，并且第一感知数据例如为摄像头数据。与第一总线相耦合的接口电路从摄像头接收原始图像或视频数据，该原始图像或视频数据例如经由第一总线传输至第一总线上的处理电路，例如AIV或AIC。处理电路对原始图像或视频数据进行处理之后获得第一感知数据。第一感知数据经由第一总线和桥接电路传输至第二总线。
在804，通过第二总线传输第二感知数据和来自于第一总线的第一感知数据，第二总线的类型与第一总线的类型不同，第二感知数据的类型与第一感知数据的类型不同。第二总线例如是环形总线，第二感知数据例如是LiDAR数据。LiDAR可以经由接口电路耦合至第一总线或第二总线。在耦合到第一总线的情形，LiDAR数据的处理和传输方式与摄像头数据相似，在此不再赘述。在LiDAR耦合到第二总线的情形下，LiDAR将原始LiDAR数据经由第二总线传输至处理电路，例如AIV或AIC，以生成第二感知数据。第二感知数据通过第二总线传输至第二总线上的处理器。在806，基于从所述第二总线获取的所述第一感知数据和所述第二感知数据生成用于控制执行装置的控制信号。处理器例如可以使用之前描述的方式基于第一感知数据和第二感知数据生成该控制信号。例如，处理器可以将第一感知数据和第二感知数据进行数据融合，以生成控制信号。在一个实施例中，处理器还可以基于第一感知数据和第一辅助数据确定第一环境数据，并且基于第二感知数据和第二辅助数据确定第二环境数据，然后将第一环境数据和第二环境数据融合以确定环境状况，并且基于所确定的环境状况生成相应的控制信号。备选地，处理器也可以基于第一环境数据和第二环境数据直接确定控制信号。
尽管已经采用特定于结构特征和/或方法逻辑动作的语言描述了本主题，但是应当理解所附权利要求书中所限定的主题未必局限于上面描述的特定特征或动作。相反，上面所描述的特定特征和动作仅仅是实现权利要求书的示例形式。

Claims (15)

  1. 一种芯片，包括：
    第一总线，被配置为传输第一感知数据；
    第二总线，耦合至所述第一总线，所述第二总线的类型与所述第一总线的类型不同，所述第二总线被配置为传输第二感知数据和来自所述第一总线的所述第一感知数据，所述第二感知数据的类型与所述第一感知数据的类型不同；以及
    处理器，耦合至所述第二总线，所述处理器被配置为基于所述第一感知数据和所述第二感知数据生成用于控制执行装置的控制信号。
  2. 根据权利要求1所述的芯片，还包括：
    第三总线，耦合至所述第一总线，所述第三总线的类型与所述第二总线相同，所述第三总线被配置为传输来自所述第一总线的与所述第一感知数据对应的第一辅助数据和与所述第二感知数据对应的第二辅助数据；
    其中所述处理器耦合至所述第三总线并且所述处理器被进一步配置为基于所述第一感知数据、所述第一辅助数据、所述第二感知数据和所述第二辅助数据生成所述控制信号。
  3. 根据权利要求2所述的芯片，其中所述处理器被进一步配置为：
    基于所述第一感知数据和所述第一辅助数据确定用于表示环境状况的第一环境数据；
    基于所述第二感知数据和所述第二辅助数据确定用于表示所述环境状况的第二环境数据；以及
    基于所述第一环境数据和所述第二环境数据生成所述控制信号。
  4. 根据权利要求3所述的芯片，其中所述处理器被进一步配置为：
    将所述第一环境数据和所述第二环境数据进行融合以生成融合数据；以及
    基于所述融合数据生成所述控制信号。
  5. 根据权利要求1所述的芯片，其中所述处理器被进一步配置为将所述第一感知数据和所述第二感知数据进行融合以生成融合数据；以及
    基于所述融合数据生成所述控制信号。
  6. 根据权利要求1-5中任一项所述的芯片，还包括：
    第一交叉站点电路，位于所述第一总线上并且被配置为传输数据；以及
    第一处理电路，耦合至所述第一交叉站点电路并且被配置为基于来自第一传感器的第一感测数据生成所述第一感知数据，所述第一传感器耦合至所述第一交叉站点电路。
  7. 根据权利要求1-6中任一项所述的芯片，还包括：
    第二交叉站点电路，位于所述第二总线上并且被配置为传输数据；以及
    第二处理电路，耦合至所述第二交叉站点电路并且被配置为基于来自第二传感器的第二感测数据生成所述第二感知数据，所述第二传感器耦合至所述第二交叉站点电路。
  8. 根据权利要求1-7中任一项所述的芯片，其中所述第一总线包括网状总线，以及
    所述第二总线包括环形总线。
  9. 一种电子组件，包括：
    电路板；以及
    根据权利要求1-8中任一项所述的芯片，所述芯片被安装在所述电路板上。
  10. 一种交通工具，包括：
    电源；以及
    根据权利要求9所述的电子组件，所述电子组件由所述电源供电。
  11. 根据权利要求10所述的交通工具，还包括：
    第一传感器，经由第一接口耦合至第一总线并且被配置为生成第一传感器数据并且将所述第一传感器数据传输至所述第一总线；以及
    第二传感器，经由第二接口耦合至第二总线，并且配置为生成第二传感器数据并且将所述第二传感器数据传输至所述第二总线。
  12. 一种用于生成控制信号的方法，包括：
    通过第一总线传输第一感知数据；
    通过第二总线传输第二感知数据和来自于所述第一总线的所述第一感知数据，所述第二总线的类型与所述第一总线的类型不同，所述第二感知数据的类型与所述第一感知数据的类型不同；
    基于从所述第二总线获取的所述第一感知数据和所述第二感知数据生成用于控制执行装置的控制信号。
  13. 根据权利要求12所述的方法，还包括：
    通过第三总线传输来自所述第一总线的与所述第一感知数据对应的第一辅助数据和与所述第二感知数据对应的第二辅助数据，所述第三总线耦合至所述第一总线，所述第三总线的类型与所述第二总线相同；以及
    基于所述第一感知数据、所述第一辅助数据、所述第二感知数据和所述第二辅助数据生成所述控制信号。
  14. 根据权利要求13所述的方法，其中基于从所述第二总线获取的所述第一感知数据和所述第二感知数据生成用于控制执行装置的控制信号包括：
    基于所述第一感知数据和所述第一辅助数据确定用于表示环境状况的第一环境数据；
    基于所述第二感知数据和所述第二辅助数据确定用于表示所述环境状况的第二环境数据；以及
    基于所述第一环境数据和所述第二环境数据生成所述控制信号。
  15. 根据权利要求14所述的方法，其中基于所述第一环境数据和所述第二环境数据生成所述控制信号包括：
    将所述第一环境数据和所述第二环境数据进行融合以生成融合数据；以及
    基于所述融合数据生成所述控制信号。
PCT/CN2021/094166 2021-05-17 2021-05-17 用于交通工具的集成电路芯片 WO2022241620A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
PCT/CN2021/094166 WO2022241620A1 (zh) 2021-05-17 2021-05-17 用于交通工具的集成电路芯片
CN202180093275.2A CN116888927A (zh) 2021-05-17 2021-05-17 用于交通工具的集成电路芯片
EP21940077.7A EP4325789A4 (en) 2021-05-17 2021-05-17 INTEGRATED CIRCUIT CHIP FOR A VEHICLE
US18/510,972 US20240089143A1 (en) 2021-05-17 2023-11-16 Integrated circuit chip for transportation vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/094166 WO2022241620A1 (zh) 2021-05-17 2021-05-17 用于交通工具的集成电路芯片

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US18/510,972 Continuation US20240089143A1 (en) 2021-05-17 2023-11-16 Integrated circuit chip for transportation vehicle

Publications (1)

Publication Number Publication Date
WO2022241620A1 true WO2022241620A1 (zh) 2022-11-24

Family

ID=84141006

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/094166 WO2022241620A1 (zh) 2021-05-17 2021-05-17 用于交通工具的集成电路芯片

Country Status (4)

Country Link
US (1) US20240089143A1 (zh)
EP (1) EP4325789A4 (zh)
CN (1) CN116888927A (zh)
WO (1) WO2022241620A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104753551A (zh) * 2013-12-31 2015-07-01 ***通信集团公司 一种车载终端设备、行车用量数据的采集方法以及***
CN105527959A (zh) * 2015-12-29 2016-04-27 广州汽车集团股份有限公司 整车数据检测***以及方法
CN209117870U (zh) * 2018-11-14 2019-07-16 深圳市数聚天源人工智能有限公司 智能车辆的感知***
US20210021442A1 (en) * 2019-07-16 2021-01-21 Baidu Usa Llc Open and safe monitoring system for autonomous driving platform

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102011088020A1 (de) * 2011-12-08 2013-06-13 Bayerische Motoren Werke Aktiengesellschaft Kommunikationseinrichtung für ein Kraftfahrzeug
US9710413B2 (en) * 2013-06-28 2017-07-18 Stmicroelectronics S.R.L. Integrated data concentrator for multi-sensor MEMS systems
US10601425B2 (en) * 2018-05-30 2020-03-24 Intel Corporation Width and frequency conversion with PHY layer devices in PCI-express
JP7059899B2 (ja) * 2018-11-09 2022-04-26 トヨタ自動車株式会社 ネットワークシステム

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104753551A (zh) * 2013-12-31 2015-07-01 ***通信集团公司 一种车载终端设备、行车用量数据的采集方法以及***
CN105527959A (zh) * 2015-12-29 2016-04-27 广州汽车集团股份有限公司 整车数据检测***以及方法
CN209117870U (zh) * 2018-11-14 2019-07-16 深圳市数聚天源人工智能有限公司 智能车辆的感知***
US20210021442A1 (en) * 2019-07-16 2021-01-21 Baidu Usa Llc Open and safe monitoring system for autonomous driving platform

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4325789A4 *

Also Published As

Publication number Publication date
US20240089143A1 (en) 2024-03-14
EP4325789A4 (en) 2024-07-03
CN116888927A (zh) 2023-10-13
EP4325789A1 (en) 2024-02-21

Similar Documents

Publication Publication Date Title
CN109917765B (zh) 一种基于自动驾驶***的网络架构的集散式域控制器***
CN109477723B (zh) 自动驾驶车辆运动规划和可重新配置的运动规划处理器
JP5255579B2 (ja) 車内データ中継装置、車両制御システム
US11183071B2 (en) Drone flight optimization using drone-to-drone permissioning
CN209842367U (zh) 一种基于自动驾驶***的网络架构的集散式域控制器***
US10225168B2 (en) Interface apparatus and memory bus system
CN101901205B (zh) 在PCIExpress上启用基于ID的流的方法和装置
KR20210033996A (ko) 전용 저 레이턴시 링크를 사용한 다수의 하드웨어 가속기에 대한 통합된 어드레스 공간
CN116170779B (zh) 一种协同感知数据传输方法、装置及***
TWI783094B (zh) 合作感知
KR20190046477A (ko) 게이트웨이에서의 메시지 처리 방법 및 게이트웨이
US10742740B2 (en) In-vehicle network system
JP2023024931A (ja) システム・オン・チップの分離された安全領域への障害の伝達
CN114179817A (zh) 一种车辆控制器、车辆和车辆控制方法
JP2020082855A (ja) 車両制御装置及び車両制御方法
WO2022241620A1 (zh) 用于交通工具的集成电路芯片
CN209928279U (zh) 智能驾驶***及智能车辆
Kenjić et al. Connectivity challenges in automotive solutions
US11968117B2 (en) Daisy chain network of sensors
WO1993005602A1 (en) Multiplex transmission system
WO2018198783A1 (ja) 車両制御システム検証手法および検証装置および制御装置
CN106302243A (zh) 一种报文传输方法、cpu以及网络设备
US20070189296A1 (en) Data transmission circuit and method for controlling the data transmission circuit
KR20180128576A (ko) 다중 코어 제어 방법
US20240089142A1 (en) Vehicle-mounted apparatus and a method for relaying

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 202180093275.2

Country of ref document: CN

WWE Wipo information: entry into national phase

Ref document number: 2021940077

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2021940077

Country of ref document: EP

Effective date: 20231117

NENP Non-entry into the national phase

Ref country code: DE