Distributed Time Division Multiplexing Media Access Control System and Method for Driving Safety

Abstract

A distributed time-division multiplexing control system and method for driving safety includes: a power control module for driving safety awareness, a lightweight chain structure maintenance module, and a conflict-free time slot allocation module. This invention considers that each vehicle node in the network has different safety requirements. The core idea of ​​this invention is to match the V2V communication range with the driving safety requirements of vehicles, exchanging safety messages only between vehicles that are mutually concerned about safety, thereby greatly improving the utilization rate of channel resources. Vehicles with mutual safety concerns in the same lane form a flexible and dynamic broadcast chain structure. This invention combines power control based on driving safety awareness with lightweight dynamic intra-chain role switching, which can avoid access conflicts and merging conflicts in highly dynamic vehicle networks.

Description

Technical Field

[0001] This invention relates to a technology in the field of autonomous driving, specifically a fully distributed time-division multiplexing control system and method (BubbleMAC) for driving safety. Background Technology

[0002] Existing vehicular network (V2V) designs employ broadcast media access control (MAC) methods, including centralized and distributed approaches. The former heavily relies on a central control node and cannot function without it, while also consuming significant channel resources. The latter cannot guarantee communication latency, especially in high-vehicle-density scenarios, where it suffers from hidden terminal problems leading to severe delays. Furthermore, in high-vehicle-density scenarios, it faces issues such as insufficient channel resource allocation and transmission conflicts between vehicle nodes. Summary of the Invention

[0003] This invention addresses the shortcomings of existing technologies in achieving reliable vehicle communication and operating reliably in highly dynamic, high-density vehicle scenarios. It proposes a distributed time-division multiplexing control system and method for driving safety. Considering that each vehicle node in the network has different safety needs (e.g., vehicles traveling straight are most concerned with the vehicles ahead, while vehicles changing lanes are concerned with the vehicles in front and behind the target lane), the core idea of ​​this invention is to match the V2V communication range with the driving safety needs of vehicles, exchanging safety messages only between vehicles with mutual safety concerns, thereby greatly improving the utilization of channel resources. Vehicles in the same lane with mutual safety concerns form a flexible and dynamic broadcast chain structure. This invention combines driving safety awareness-based power control and lightweight dynamic intra-chain role switching to avoid access and merging conflicts in highly dynamic vehicle networks.

[0004] This invention is achieved through the following technical solution:

[0005] This invention relates to a distributed time-division multiplexing control system for driving safety, comprising: a driving safety-aware power control module, a lightweight chain structure maintenance module, and a conflict-free time slot allocation module. The driving safety-aware power control module determines a safe distance based on local vehicle information and received neighbor vehicle information, thereby achieving communication power control based on mutual safety concerns to conserve channel resources. The lightweight chain structure maintenance module judges and handles the growth and breakage of the chain structure based on whether neighbor vehicle information is received, updating vehicle roles in real time in a highly dynamic environment. The conflict-free time slot allocation module allocates non-overlapping time slot resources to each role based on vehicle role information in the chain structure, achieving collision-free and collision-free time slot allocation, ensuring reliable safe broadcasting between vehicles.

[0006] This invention relates to a distributed time-division multiplexing media access control method (BubbleMAC) based on the above system. In the initial stage, each vehicle determines the minimum safe distance to maintain from the vehicle in front based on its own and the vehicle in front's driving status, and adaptively adjusts the communication range to cover the vehicles in front and behind based on the beacon reception of the vehicles in front and behind. A chain structure is formed between the vehicles that communicate with each other to identify the role of each vehicle in the chain and the information of the chain. The chain is updated and maintained locally by each member of the chain, and conflict-free time slots are allocated to each role in the chain structure.

[0007] Technical effect

[0008] This invention quantifies driving safety requirements, allowing vehicles to adaptively adjust their communication range based on each other's safety needs. It divides channel time slots into non-overlapping resource pools and constructs a dynamic broadcast chain structure. In this chain structure, vehicles immediately know and update their roles based on beacon reception. Compared with existing technologies, this invention limits the communication range to improve the spatial utilization of channel resources while ensuring driving safety, allocates conflict-free time slots to vehicles with different densities and mobility attributes, and maintains the chain structure at low communication costs in a highly dynamic environment. Attached Figure Description

[0009] Figure 1 This is a schematic diagram of the present invention;

[0010] Figure 2 This is a diagram showing the one-hop set (OHS) and two-hop set (THS) of vehicle n1, as well as an access conflict.

[0011] Figure 3 , 4 This is a diagram of the architecture of the present invention.

[0012] Figure 5 This is a diagram illustrating safe distances.

[0013] In the diagram: (a) indicates that n has been obtained. j Driving status information, vehicle n i Get n i With n j (a) The required safe distance between vehicles; (b) When the driving status information of the vehicle in front cannot be obtained, vehicle n j It is necessary to assume that there is a completely stopped car in front of you to calculate your safe distance;

[0014] Figure 6 This is a schematic diagram of the chain structure growth process;

[0015] In the diagram: (a) represents the approach of two chains, and (b) represents the merging of two chains;

[0016] Figure 7 This is a schematic diagram of the chain structure fracture process;

[0017] In the diagram: (a) represents members moving away from each other in the chain, and (b) represents the chain breaking.

[0018] Figure 8 This is a schematic diagram of the frame structure of the present invention;

[0019] Figure 9 A schematic diagram showing the approach of two chains on different lanes;

[0020] In the diagram: (a) represents the approach of two chains, and (b) represents the intersection of two chains;

[0021] Figure 10 This is a schematic diagram illustrating examples of the present invention in other complex driving scenarios;

[0022] In the diagram: (a) represents a lane-changing scenario, where vehicle n1 waits and changes lanes to turn right; (b) represents an intersection scenario, where vehicle n2 waits for vehicle n1 to pass straight before turning right.

[0023] Figure 11 This is a schematic diagram of the MAC layer data packet structure of this media access control method;

[0024] Figure 12 This is a schematic diagram of the road topology in the simulation experiment;

[0025] In the diagram: (a) is a highway, (b) is a three-way intersection, and (c) is a four-way intersection;

[0026] Figure 13 This is a schematic diagram illustrating the impact of different vehicle densities on driving safety in a simulation experiment.

[0027] In the figure: (a) is the transmission collision rate, (b) is the packet reception rate, and (c) is the required packet reception rate;

[0028] Figure 14 This is a schematic diagram illustrating the impact of different road topologies on driving safety in a simulation experiment.

[0029] In the figure: (a) is the transmission collision rate, (b) is the packet reception rate, and (c) is the required packet reception rate;

[0030] Figure 15 This is a schematic diagram illustrating the impact of abnormally moving vehicles on driving safety under various vehicle densities during a simulation experiment.

[0031] In the figure: (a) represents low traffic density, (b) represents medium traffic density, and (c) represents high traffic density;

[0032] Figure 16 This is a schematic diagram showing the distribution of the slot set |OHN| in a one-hop distribution and the slot set |Chain| within a chain structure.

[0033] In the figure: (a) is the |OHN| distribution, and (b) is the Chain distribution. Detailed Implementation

[0034] like Figure 1 As shown, vehicles adaptively adjust their communication range based on each other's safety concerns. Vehicles n1 (lead vehicle), n2 (middle vehicle), and n3 (follower vehicle) in the same lane form a chain structure. Within this chain, each vehicle is allocated a conflict-free time slot for sharing and forwarding critical driving status and decision-making information. When one chain structure approaches another in the same direction, lightweight negotiation occurs, and the approaching chain structure makes necessary time slot changes to avoid potential message merging conflicts.

[0035] like Figure 3 As shown in this embodiment, a distributed time-division multiplexing media access control system for driving safety includes: a driving safety-aware power control module, a lightweight chain structure maintenance module, and a conflict-free time slot allocation module. The driving safety-aware power control module determines a safe distance based on local vehicle information and received neighbor vehicle information, thereby achieving communication power control based on mutual safety concerns to save channel resources. The lightweight chain structure maintenance module judges and handles the growth and breakage of the chain structure based on whether neighbor vehicle information is received, updating vehicle roles in real time in a highly dynamic environment. The conflict-free time slot allocation module allocates non-overlapping time slot resources to each role based on the vehicle role information in the chain structure, achieving collision-free and collision-free time slot allocation, ensuring reliable and safe broadcasting between vehicles.

[0036] The time slot resource refers to the following: In a time-division multiplexing-based media access control (MAC) method, time is divided into a series of frames, each consisting of a fixed number of time slots. All vehicles synchronize via GPS, and each vehicle is assigned a time slot to transmit messages. Once a vehicle node successfully acquires a time slot, it will occupy the same time slot in all subsequent frames until a transmission collision is detected. Figure 2 As shown, neighboring vehicles within a vehicle's communication range are considered as the one-hop set of that vehicle node. The union of the nodes of the OHS of each neighbor node in the vehicle's OHS is the two-hop set (THS) of that vehicle, i.e., ∪ for When n1 and n2 ∈ THS(n1) select the same time slot, an access conflict occurs at n5. When a distant vehicle node n... j Vehicle node n is added due to relative motion. i When THS devices share the same time slot, a merge conflict will occur.

[0037] The chain structure refers to a situation where two or more vehicles in the same lane are in a mutual safety concern relationship, and each vehicle is at the minimum safe distance d from the vehicle behind it. safe When the vehicle is inside, a chain structure is formed, in which each vehicle adjusts its communication range to cover the vehicles in front and behind.

[0038] Due to the lack of centralized coordination, two types of message conflicts may occur: access conflicts and merge conflicts. When vehicle node n... i and THS(n i Another vehicle node n in ) j Access conflicts occur when the same time slot is selected within a frame. For example, in Figure 2 In the context of THS (n1), if n1 and n6 ∈ THS (n1) select the same time slot, an access conflict occurs at n5; when the distant vehicle node n j Vehicle node n is added due to relative motion. i When vehicles share the same time slot within their respective HHS sets, merging conflicts can occur. To avoid access conflicts, each vehicle can include information about its neighbors' occupied time slots in its beacon set. After listening to a frame on the channel, a vehicle can collect information about its neighbors' occupied time slots in its HHS and select an available time slot for all subsequent frames. However, due to the diverse mobility of vehicles, merging conflicts are difficult to eliminate in a fully distributed environment.

[0039] The power control includes:

[0040] Step a) Determine the safe distance: such as Figure 5 As shown in (a), when vehicle n i With t beac. The time interval (e.g., 100ms) is communicated with n via a safety beacon. j Exchange their driving status information. When n j Suddenly brake at maximum deceleration, n i After receiving n j Braking begins after the beacon is sighted. To avoid a collision, n i It should be at least with n j Maintain a safe distance Where: v i and v j They are n i and n j speed; b i and b j They are n i and n j Maximum deceleration; ρ i is n i Braking coefficient adopted; T reac.This refers to the reaction time of a human driver (typically 0.6-0.8 seconds) or the reaction time of an autonomous driving agent. Considering the worst-case scenario, such as... Figure 5 As shown in (b), n i When it cannot hear the beacons of other vehicles, it needs to assume there is a completely stationary vehicle ahead and calculate the maximum safe distance.

[0041] Therefore, to ensure driving safety, n i Its transmission power should be controlled to ensure that its communication range is greater than d. safe Furthermore, when n i When the speed is 0, the safe distance can be calculated from the above formula to be 0. At this time, n i Stop active broadcasting.

[0042] Step b) Power control based on mutual safety concerns: As can be seen from the safety distance calculation, n i of It is proportional to the square of the speed. When a following car with a higher speed n k Approaching n i Obviously n k Maximum safe distance $r = n i Large. When n i Located at n k When within the communication range, n i You can hear n k The beacon is sent, but the reverse is not true. In this case, when n... i Brake, n k No notification will be received. This invention refers to $n. i and n k With mutual security concerns, n i Its communication range should be expanded to cover its own d safe and following vehicles with mutual safety concerns safe For other driving scenarios, such as changing lanes or turning at intersections, this can be achieved by equivalently... i The vehicle has been moved to the target lane to switch to a straight-ahead scenario.

[0043] The aforementioned lightweight chain structure maintenance refers to: when vehicle n i When establishing communication links with other vehicles in the same lane that have safety concerns, a dynamic and flexible communication chain structure is formed. Based on n i Whether a two-way link can be established between the vehicle and the vehicle in the chain structure, allowing the vehicle to dynamically switch roles and cause the chain structure to grow and break, makes the growth and breakage process of the chain structure very lightweight.

[0044] The conflict-free time slot allocation refers to the following: time slots in each frame of time-division multiplexing are divided into disjoint sets for communication between vehicles with different roles. To avoid acquisition conflicts, each vehicle node adds the time slot usage information of nodes in its OHS set and the time slot usage information of nodes in its chain structure as control information to the data packet. Nodes select available time slots for use by collecting the time slot information occupied by their neighbors in their THS and nodes within the chain. To avoid merging conflicts, during the expansion or splitting of the chain structure, when a vehicle role changes, the vehicle selects a new time slot from the corresponding dedicated time slot set. When two chain structures in different lanes meet, the lead vehicle of the closer chain structure coordinates the time slot usage of its chain members to avoid using time slots already occupied in the preceding chain structure.

[0045] like Figure 4 As shown, this embodiment relates to a fully distributed time-division multiplexing control method for driving safety based on the aforementioned system. In the initial stage, each vehicle determines the minimum safe distance to the vehicle in front based on its own and the vehicle ahead's driving status. It then adaptively adjusts its communication range to cover both vehicles based on the beacon reception status of the preceding and following vehicles. A chain structure is formed between the mutually communicating vehicles to identify each vehicle's role within the chain and the chain's information. Updates and maintenance are performed locally at each chain member, and conflict-free time slot allocation is performed for each role in the chain structure. Specifically, this includes:

[0046] Step 1) Initially, the vehicle broadcasts its local information in single-vehicle mode, determining its own driving status and braking preferences based on the vehicle's driving state and braking preferences. i With the assumed completely stationary vehicle node n j Minimum safe distance d between safe The communication range is limited by adjusting the transmission power to ensure that it covers at least the minimum safe distance d. safe When this car n i Received from the preceding vehicle n j Or receive a signal from a vehicle behind in the same lane. k At its minimum safe distance d safe When sending beacons within range, this vehicle n i Adjust its communication range to cover the vehicle in front n j Or the following car n k , and form a chain structure;

[0047] Step 2) Lightweight chain structure maintenance is performed distributed among all vehicles within the chain structure. When a new chain structure is generated through merging or an existing chain structure breaks, each vehicle quickly infers its relative position in the chain based on the received beacon to determine its role.

[0048] Step 3) After determining the role of this vehicle, each vehicle in the chain structure is allocated non-overlapping time slot resources according to its role. The process mainly includes the allocation of time slots without access conflicts and the reallocation of time slots without merging conflicts, thereby achieving reliable and conflict-free secure broadcasting.

[0049] The vehicle roles include: isolated vehicle, head vehicle, intermediate vehicle, and tail vehicle. An isolated vehicle, i.e., a single vehicle, is one that cannot establish a communication link with a neighboring vehicle with mutual safety concerns in the same lane; this means it does not form a chain. A head vehicle (or tail vehicle) is one that can form a chain with a neighboring vehicle with mutual safety concerns in the forward (or backward) direction in the same lane, but cannot form a chain with vehicles in the backward (or forward) direction. An intermediate vehicle is one that can establish a two-way communication link with neighbors with mutual safety concerns in both the forward and backward directions in the same lane.

[0050] The identification of each vehicle's role within the chain refers to: confirming the role of each vehicle within the chain. i Whether a vehicle can receive beacons from vehicles in the same lane and establish a two-way link with its neighbors in both directions (front and rear) determines its role and causes the chain structure to grow or break. Chain structure growth refers to the process of adding new nodes to the chain, including: isolated vehicles encountering other isolated vehicles, isolated vehicles encountering chain structures, chain structures encountering isolated vehicles, and two chain structures meeting. Chain structure breaking occurs when the distance between two vehicles in the chain gradually increases due to speed differences until it exceeds the distance d of the following vehicle. safe At that time, the chain structure broke.

[0051] like Figure 6 As shown, n1 and n2 are the middle and rear carriages of the slower chain. n3 and n4 are the front and middle carriages of the faster chain. When the faster chain encounters the slower chain, the faster chain receives the slot allocation information from the slower chain, and the two chains merge into a new chain. Figure 6 As shown in (a), when a faster chain i approaches a slower chain j in the same lane, the tail car n2 of j will be within a safe distance of the head car n3 of i. In this case, chain growth occurs. Specifically, after n2 receives a beacon from n3, it adjusts its communication range to cover n3 and its preceding neighbor n1. Similarly, after n3 receives a beacon from n2, it adjusts its communication range to cover n2 and n4. Finally, both n2 and n3 change their roles, merging chain i into chain j. If n2 (or n3) was an isolated vehicle before merging, it will become the head (or tail) car of chain j. Otherwise, as... Figure 6As shown in (b), n3 and n2 both become intermediate vehicle nodes of chain j, and the slot occupancy is updated according to the channel resource status below.

[0052] As the distance between two vehicles in the chain gradually increases due to the speed difference, it eventually exceeds the distance d between the following vehicles. safe When the chain structure breaks, specifically, this includes: breaking from the lead vehicle, breaking from the middle vehicle, and breaking from the tail vehicle on the same lane. Specifically: in the case of a break from the lead or tail vehicle, the original lead or tail vehicle in the chain becomes a single vehicle, and the original middle vehicle becomes the lead or tail vehicle of the new chain; in the case of a break from the middle vehicle, the two disconnected vehicle nodes become the tail and lead vehicles of two new chain structures, respectively. For example... Figure 7 As shown, the distance between two vehicles in the same chain gradually increases until it exceeds the safe distance d for the following vehicles. safe At that time, the chain structure will break into two chains;

[0053] like Figure 7 As shown in (a), the chain structure breaks when the distance between a pair of members n2 and n3 in chain i gradually increases until it exceeds the maximum safe distance of n3. Specifically, when n2 cannot receive a beacon from n3, it considers the chain to have broken. Figure 7 As shown in (b), at this time n2 will adjust its communication range to only cover the preceding vehicle n1, and change its role from the middle vehicle node (or the first vehicle node) of chain i to the last vehicle node (or the isolated vehicle node) of chain i.

[0054] The aforementioned non-overlapping time slot resource allocation refers to the following: to avoid data transmission conflicts, vehicles select time slots based on their roles in the chain structure. For example... Figure 8 As shown, the time slot resources for each frame are divided into three disjoint sets: L for nodes in the left direction, R for nodes in the right direction, and F for fixed roadside units (RSUs). L and R are further divided into four subsets: H for leading vehicle nodes, I for middle vehicle nodes, T for trailing vehicle nodes, and S for isolated vehicle nodes. This divides the frame into three disjoint time slot sets L, R, and F. F is reserved for RSUs, and L and R are used by vehicles on the left and right sides of the road, respectively. Here, left and right are defined as road segments extending from south / north to west (east) being considered left or right road segments, as shown in the diagram. Figure 8 As shown. The resource allocation into L and R is designed to address drastic changes in the relative positions of vehicles in opposing lanes. To reduce potential conflicts caused by changes in the relative positions of vehicles traveling in the same direction, both time slot resource sets L and R are further divided into H, I&S, and T, where: H is used by the lead vehicle in the chain, T is used by the tail vehicle in the chain, and I&S is used by the middle vehicles and isolated vehicles in the chain.

[0055] The collision-free slot allocation refers to eliminating access and merging collisions during packet transmission by utilizing disjoint slot sets and dynamic changes in intra-chain roles. Vehicle node ni not only broadcasts its own driving status and decisions but also shares slot usage information for vehicles in the OHS set and their respective chain structures. Vehicles on the road should first listen to the channel (for one frame) to determine the slots occupied by their THS node and nodes in their chain structure (when not a single vehicle), and determine the set of unoccupied slots O. Vehicles randomly select a slot from the intersection of the available slot set and their corresponding set for access. For example, the lead vehicle in the left lane should randomly select a slot from O∩H∩L. When a vehicle in the same lane does not hear any packets from its neighbors, it is considered an isolated vehicle and should randomly select a slot from O∩I&S∩L (or O∩I&S∩R) when it is in the left (right) lane. Because vehicles with different roles obtain slots from disjoint slot resource pools, access collisions are almost completely eliminated. Collisions can only occur when vehicles with the same role simultaneously acquire the same time slot. Specifically, when an isolated vehicle in one lane encounters another isolated vehicle, the probability of them choosing the same time slot is finite, since they both randomly select time slots. When there are k available time slots, the probability of them using the same time slot in the nth consecutive frame is...

[0056] The aforementioned collision-free time slot allocation includes:

[0057] a) Scenario on the same lane. This invention takes the encounter of two chain structures on the same lane as an example. For example... Figure 6 As shown, n3 and n4 are the lead and middle vehicles of the faster chain, respectively, while n1 and n2 are the middle and tail vehicles of the slower chain, respectively. Before the chain structure completes its growth, n3 always acts as the lead vehicle, and n2 always acts as the tail vehicle. Since the time slot sets of different roles do not intersect, they will always use different time slots, thus preventing transmission collisions during the growth process. During the approach, n3 obtains the time slot allocation of the slower chain ahead from n2 and adds this information as control information to the broadcast data packet. Other vehicles in the faster chain adjust their broadcast time slots according to the platoon's time slot information occupancy. The other three growth scenarios and the disconnection scenario are handled in a similar manner.

[0058] b) Scenario involving different lanes. Vehicles in different lanes cannot form a chain structure, nor can they merge as in the scenario described above. This invention uses a lightweight negotiation mechanism to eliminate merging transmission conflicts caused by the relative motion of vehicles in different lanes. For example... Figure 9As shown, n3 and n4 are the lead and middle trains of the faster chain. n1 and n2 are the middle and tail trains of the slower chain. When the faster chain i intersects with another chain j in a different lane, their respective chain structures do not change. However, the lead train n3 of chain i is responsible for reallocating time slots based on the channel usage of chain j. Specifically, when n3 receives a beacon from the tail train n2 of chain j, it learns about all members in chain j and their used time slots. Similar to the chain merging situation in the same lane, n3 coordinates all members in chain i to select unused time slots in chain j. For example, n4 is notified to select another unused time slot to avoid a conflict with n1 in chain j.

[0059] c) Intersection Scenario. Intersections are challenging for MAC control methods due to high vehicle density and the inability of vehicles from different directions to coordinate. This invention adjusts broadcast behavior based on traffic lights to eliminate merging collisions at intersections. Specifically, at an intersection, when an isolated vehicle or the leading vehicle stops, it should turn off its radio and notify following vehicles. Even if it doesn't broadcast, following vehicles will calculate d safe There is still a stationary vehicle present. For example, when the north-south green light is on and the east-west red light is on, the behavior of vehicles going straight is the same as in a non-intersection scenario with different lanes. At this time, stationary vehicles in lanes 1 and 5 turn off their radios to avoid interfering with broadcasts from north-south vehicles. If a vehicle in the right-turn lane of lanes 1 and 5 wants to turn right at this time, it will join the main road traffic as an isolated vehicle, i.e., it becomes a situation of an isolated vehicle encountering a chain structure or an isolated vehicle encountering another isolated vehicle.

[0060] like Figure 10 As shown in (a), when n1 intends to change lanes to the right lane, if n1 can hear the beacon sent by n2, it can calculate whether it is safe for n1 to change lanes for n2 based on the current driving state of the neighboring nodes. If so, n1 gains the right-of-way and takes appropriate action to obtain a new role and time slot in the target lane; otherwise, n1 must wait for a suitable opportunity (e.g., accelerate to a safe distance d2). Another example is... Figure 10 As shown in (b), a right-turning vehicle n2 yields to a straight-going vehicle n1. Although they both select a slot from the R set, no merging conflict occurs because when n2 stops at the intersection, its speed is 0, and it stops broadcasting due to power control of traffic safety awareness. However, since n2 is still listening to the channel, it can determine when it is safe to turn right and obtain a new role and slot on the target road segment. Therefore, just as vehicles strictly adhering to traffic rules will not cause physical collisions, vehicles broadcasting in accordance with this media access control method will not experience message conflicts.

[0061] like Figure 11As shown, the data packet (MSDU) includes: an 802.11 MAC header and a MAC layer data unit (MAC Service Data Unit), wherein: the first byte represents the one-hop set of vehicle n1. The size, followed by each member The first byte contains the vehicle ID and the obtained time slot; the next byte indicates the size of chain i, followed by the vehicle ID and the obtained time slot of each member in chain i; the remaining fields are motion information and driving decision information such as the current lane number, speed, longitude, latitude, and maximum deceleration.

[0062] Through specific implementation, simulations are performed on a trace-driven basis. This involves using SUMO to generate vehicle trajectories for various scenarios and using the VENUS simulator to simulate different control methods. Three typical scenarios are considered: a highway, a three-way intersection, and a four-way intersection. Each road is a two-way eight-lane road with a length of 4 kilometers. Figure 12 As shown. The speed limits for the four lanes traveling in the same direction are 60 km / h, 80 km / h, 100 km / h, and 120 km / h respectively. Each intersection has traffic lights, with the green light lasting 20 seconds. There are ten different vehicle motion parameters, including acceleration (from 1 m / s). 2 up to 5m / s 2 ), deceleration capability (from 3m / s 2 Up to 10m / s 2 The simulation included maximum speeds (from 80 km / h to 240 km / h). To simulate different traffic conditions throughout the day, different traffic flow levels were set, such as high-density traffic (10 vehicles per lane per minute), medium-density traffic (5 vehicles per lane per minute), and low-density traffic (3 vehicles per lane per minute). Each vehicle randomly selected its destination and route upon entering the road and was driven using the Krauss vehicle following model and the LC2013 lane-changing model. The maximum communication range was set to 300 meters because 802.11p-compatible onboard units can reliably transmit data within 300 meters. For each simulation round, a 100-second trajectory was selected after a stable traffic flow had formed. Then, using the interactive simulation function of the VENUS simulator (which allows vehicle decision-making information to come solely from vehicle-to-everything (V2X) communication data packets), in a highway scenario with the same experimental parameters, vehicle motion was dynamically adjusted based on communication results. Emergency braking with maximum deceleration at 5%, 10%, and 15% of vehicles was included to examine the invention's effect on enhancing driving safety in extreme scenarios. Furthermore, the invention was evaluated on the real-world HighD dataset, which was collected using drones on German highways, recording trajectory data from over 110,500 vehicles.

[0063] In communication simulation experiments on real road datasets, the communication performance metrics of this method include: Message Collision Rate (MCR): the average number of transmission collisions per vehicle per frame; Beacon Reception Ratio (BRR): the average number of packets actually received per vehicle per frame divided by the expected number of packets; Utility of Safety Beacons (USB): the average number of mutual safety concern packets actually received per vehicle per frame divided by the expected number of mutual safety concern packets; Number of Collisions: the number of collisions that occurred during the simulation. During the VENUS interactive simulation, this invention considers the impact of different control methods on safe driving, primarily focusing on the number of accidents (Number of Crashes), i.e., the number of traffic accidents that occurred.

[0064] The present invention was compared with IEEE 802.11p, VeMAC and SCMAC respectively. According to the requirements of safe driving application, the broadcast frequency of all control methods was set to 10Hz, that is, 10 times per second.

[0065] The simulation experiment first evaluated the impact of traffic flow density on the performance of different control methods. Specifically, on a two-way, eight-lane, four-kilometer-long highway, three different traffic densities were used to generate traffic flow: high density (10 vehicles / minute / lane), medium density (5 vehicles / minute / lane), and low density (3 vehicles / minute / lane). The speed limits for the four lanes in each direction of the highway were 120, 100, 80, and 60 km / h, respectively. For each traffic flow density, the experiment was repeated 20 times, and the average results were calculated.

[0066] like Figure 13 As shown, the MCR of this method is very low for low, medium, and high traffic densities, at 0.005, 0.027, and 0.061 collisions / frame / vehicle, respectively. All collisions are random collisions generated when isolated vehicles access the network. In contrast, VeMAC cannot avoid merging collisions, while SCMAC and 802.11p cannot avoid both access and merging collisions, resulting in a significant increase in MCR. For example, for low, medium, and high traffic densities, VeMAC has 168 times, 74 times, and 120 times more beacon collisions than this method, respectively. Furthermore, for high traffic flow density, the BRR of this method, VeMAC, SCMAC, and 802.11p are 98.9%, 81.3%, 68.6%, and 47.9%, respectively, while USB's are 98.9%, 83.7%, 68.0%, and 63.9%.

[0067] The simulation experiment then evaluated the impact of road topology. This experiment tested the robustness of the proposed method on real vehicle trajectories (highD) and three typical road topologies: a four-kilometer-long, eight-lane, two-way highway, a three-way intersection, and a four-way intersection. The traffic light phase duration was fixed at 20 seconds. This experiment generated vehicles with moderate traffic density and kept other experimental settings similar to those described earlier.

[0068] like Figure 14 As shown, across all road topologies, VeMAC, SCMAC, and 802.11p exhibit very high average MCRs due to the concentration of vehicles with conflict slots at intersections. In contrast, our proposed method maintains extremely low MCRs on highways, three-way intersections, four-way intersections, and the highD dataset, at 0.027, 0.006, 0.004, and 0.017 collisions / frame / vehicle, respectively. Furthermore, our proposed method boasts the highest BRR and USB across all road topologies. For example, at four-way intersections, the BRRs for our proposed method, VeMAC, SCMAC, and 802.11p are 99.3%, 63.9%, 49.8%, and 62.1%, respectively, while those for USB are 99.7%, 77.8%, 67.2%, and 76.8%. In comparison, the slot access mechanism of this invention ensures conflict-free performance, making it a control method applicable to a wide range of topologies.

[0069] The simulation experiment ultimately impacted driving safety. This experiment investigated whether driving safety could be achieved by relying solely on received beacons as an information source to control vehicles. Specifically, this experiment generated vehicles with three different traffic flow densities in the highway road topology, consistent with the settings in Experiment 5.5.2. Furthermore, this experiment randomly selected 5% to 15% of anomalous vehicles, continuously braking at maximum deceleration and accelerating to their maximum speed throughout the simulation, then checking whether a collision occurred due to a lack of awareness of changes in the motion of neighboring vehicles. For each simulation setting, the experiment was repeated twenty times.

[0070] Figure 15 The average number of collisions was plotted using different media access control methods. Since our method achieves near-collision-free safety beacon broadcasting in all scenarios, no collisions occurred in any simulations. VeMAC, SCMAC, and 802.11p resulted in a considerable number of collisions even under medium traffic flow densities. In summary, all control methods reduce vehicle collisions to some extent. Our invention performs best in all scenarios, achieving zero collisions. Even in the presence of abnormal vehicles, our invention can still avoid collisions through reliable communication with neighboring vehicles.

[0071] Compared to 802.11p, the main communication overhead of VeMAC and this invention lies in coordinating the control information required for media access, including the vehicle IDs and corresponding time slot indices of neighboring vehicles in the OHS set. Let |OHN|max be the maximum number of vehicles that can exist in the vehicle's OHS set. The number of bits required to represent the vehicle ID in its THS is... Among the symbols This represents the floor function. To identify a specific time slot among s time slots, it is necessary to... Therefore, VeMAC and this invention share the same communication overhead (in bits). for:

[0072] Furthermore, the vehicles in this invention need to include chain structure information in the data packet. Similarly, when the number of vehicle nodes in the chain is |Chain| max The overhead required to maintain the time slot usage information of the chain structure is:

[0073] Therefore, the total coordination overhead of this invention is:

[0074] like Figure 16 (a) Shows the |OHN| distribution for different control methods. From this, the cost of VeMAC can be estimated. Approximately 150 bytes. This invention... The overhead is approximately 95 bytes. Figure 16 (b) Shows the distribution of |Chain| in this invention. From this, the structure of the invention can be estimated. The overhead is approximately 32.5 bytes. Since the application data size broadcast by security applications in connected vehicles is typically small, around 200-500 bytes, adding an extra 100 bytes of coordination data to the broadcast packet is acceptable, as the total packet size is much smaller than the size of the MAC layer control method data unit.

[0075] Compared to existing technologies, this method successfully restricts V2V communication between vehicles with mutual safety concerns and avoids communication problems such as merging conflicts that cannot be resolved by existing methods by reallocating time slots. Therefore, this method maximizes channel utilization and road space utilization while meeting driving safety requirements. A prototype system was implemented and extensive trajectory-based simulation experiments were conducted. The implementation of this invention is simple and has low hardware requirements. Furthermore, in all settings, the media access control method of this invention achieves near-zero collision rate message passing.

[0076] The above-described specific implementations can be partially adjusted by those skilled in the art in different ways without departing from the principles and purpose of the present invention. The scope of protection of the present invention is defined by the claims and is not limited to the above-described specific implementations. All implementation schemes within the scope of the claims are bound by the present invention.

Claims

1. A distributed time-division multiplexing control system for driving safety, characterized in that, include: The system includes a power control module for driving safety perception, a lightweight chain structure maintenance module, and a conflict-free time slot allocation module. Specifically: the power control module determines a safe distance based on local vehicle information and received neighbor vehicle information, thereby achieving communication power control oriented towards mutual safety concerns to conserve channel resources; the lightweight chain structure maintenance module judges and handles the growth and breakage of the chain structure based on whether neighbor vehicle information is received, updating vehicle roles in real time in a highly dynamic environment; and the conflict-free time slot allocation module allocates non-overlapping time slot resources to each role based on vehicle role information in the chain structure, achieving time slot allocation without access collisions and without merging collisions, ensuring reliable safety broadcasting between vehicles. The aforementioned chain structure refers to a situation where two or more vehicles in the same lane are in a mutual safety concern relationship, and each vehicle is at the minimum safe distance from the vehicle behind it. When inside, a chain structure is formed, in which each vehicle adjusts its communication range to cover the vehicles in front and behind; The power control includes: Step a) Determine a safe distance: When vehicles by The time interval is transmitted via a safety beacon and Exchange their driving status information, when Suddenly brake at maximum deceleration. Upon receiving Braking was initiated after the beacon was sighted to avoid a collision. Should at least with Maintain a safe distance and They are and speed; and They are and The maximum deceleration; yes The braking coefficient adopted; This refers to the reaction time of a human driver or the reaction time of an autonomous driving agent, considering the worst-case scenario. When it cannot hear the beacons of other vehicles, it needs to assume there is a completely stationary vehicle ahead and calculate the maximum safe distance. , Its transmission power should be controlled to ensure its communication range is greater than [a certain value]. In addition, when When the speed is 0, the safe distance can be calculated from the above formula to be 0. Stop active broadcasting; Step b) Power control based on mutual safety concerns: As can be seen from the safety distance calculation, of It is proportional to the square of the speed, when a following car with a higher speed... Approaching Obviously Maximum safe distance Compare big, when lie in When within the communication range, Hear The beacon is sent, but the reverse is not true; in this case, when brake, You will not be able to receive notification, that is and Mutual security concerns Its communication range should be expanded to cover its own area. and following vehicles with mutual safety concerns For other driving scenarios, by equivalently when The vehicle has been moved to the target lane to transition to a straight-ahead scenario. The conflict-free time slot allocation refers to the following: time slots in each frame of time-division multiplexing are divided into non-overlapping sets for communication between vehicles with different roles. To avoid acquisition conflicts, each vehicle node adds the time slot usage information of nodes in its one-hop set OHS and the time slot usage information of nodes in its chain structure as control information to the data packet. Nodes select available time slots for use by collecting the time slot information occupied by their neighbors in their two-hop set THS and nodes in the chain structure. To avoid merging conflicts, when a vehicle role changes during the expansion or splitting of the chain structure, the vehicle will select a new time slot from the corresponding dedicated time slot set. When two chain structures in different lanes meet, the lead vehicle of the closer chain structure will coordinate the time slot usage of its chain members to avoid using time slots that are already occupied in the chain structure ahead.

2. The distributed time-division multiplexing control system for driving safety according to claim 1, characterized in that, The aforementioned lightweight chain structure maintenance refers to: when the vehicle When establishing communication links with other vehicles in the same lane that are of safety concern, a dynamic and flexible communication chain structure is formed, based on... Whether a two-way link can be established between the vehicle and the vehicle in the chain structure, allowing the vehicle to dynamically switch roles and cause the chain structure to grow and break, makes the growth and breakage process of the chain structure very lightweight.

3. A distributed time-division multiplexing media access control method based on the system described in claim 1 or 2, characterized in that, In the initial stage, each vehicle determines the minimum safe distance to the vehicle in front based on its own and the vehicle in front's driving status, and adaptively adjusts the communication range to cover the vehicles in front and behind based on the beacon reception of the vehicles in front and behind. A chain structure is formed between the vehicles that communicate with each other to identify the role of each vehicle in the chain and the information of the chain. The chain is updated and maintained locally by each member of the chain, and the role in the chain structure is allocated without conflict.

4. The method according to claim 3, characterized in that, specifically include: Step 1) Initially, the vehicle broadcasts its local information in single-vehicle mode, determining its own driving status and braking preferences. Compared to the assumed completely stationary vehicle node in front Minimum safe distance between The communication range is limited by adjusting the transmission power to ensure that it covers at least the minimum safe distance. When this vehicle Received the vehicle in front Or receive a vehicle behind in the same lane At its minimum safe distance When sending beacons within range, this vehicle Adjust its communication range to cover the vehicle in front. or the following vehicle , and form a chain structure; Step 2) All vehicles in the chain structure are distributed to perform lightweight chain structure maintenance. When a new chain structure is generated by merging or an existing chain structure is broken, each vehicle can quickly infer its relative position in the chain based on the received beacon to determine its role. Step 3) After determining the role of this vehicle, each vehicle in the chain structure is allocated non-overlapping time slot resources according to its role. The process mainly includes the allocation of time slots without access conflicts and the reallocation of time slots without merging conflicts, thereby achieving reliable and conflict-free secure broadcasting. The vehicle roles mentioned include: isolated vehicles, lead vehicles, middle vehicles, and tail vehicles. An isolated vehicle, i.e., a single vehicle, is a vehicle that cannot establish a communication link with a neighboring vehicle with mutual safety concerns in the same lane, which means that it does not form a chain structure. A lead or tail vehicle is a vehicle that forms a chain structure with a neighboring vehicle with mutual safety concerns in the front or rear directions in the same lane, but cannot form a chain structure with vehicles in the rear or front directions. A middle vehicle is a vehicle that establishes a two-way communication link with a neighboring vehicle with mutual safety concerns in the front and rear directions in the same lane. The data packet includes: a MAC header and a MAC layer data unit (MAC service data unit), wherein: the first byte indicates the vehicle... A set of jumps The size, followed by each member The first byte contains the vehicle ID and the obtained time slot; the next byte indicates the size of chain i, followed by the vehicle ID and the obtained time slot of each member in chain i; the remaining fields are the current lane number, speed, longitude, latitude, maximum deceleration, and driving decision information.

5. The method according to claim 3, characterized in that, The aforementioned identification of each vehicle's role within the chain refers to: confirming the vehicle's role within the chain. Whether a vehicle can receive beacons from vehicles in the same lane and establish a two-way link with its neighbors in both directions (front and rear) determines its role and causes the chain structure to grow or break. Chain structure growth refers to the process of adding new nodes to the chain, including: isolated vehicles encountering other isolated vehicles, isolated vehicles encountering chain structures, chain structures encountering isolated vehicles, and two chain structures meeting. Chain structure breaking occurs when the distance between two vehicles in the chain gradually increases due to speed differences until the vehicle overtakes the following vehicle. At that time, the chain structure broke.

6. The method according to claim 4, characterized in that, The aforementioned non-overlapping time slot resource allocation refers to the following: to avoid data transmission conflicts, vehicles select time slots based on their role in the chain structure, and the time slot resources of each frame are divided into three non-overlapping sets, namely... Nodes used in the left direction, Nodes used in the right direction, For use as fixed roadside units; and Further divided into four subsets, namely Used for the lead car node. Used for intermediate vehicle nodes. Used for the tail car node For isolated vehicle nodes, the frame is divided into three non-overlapping time slot sets. , and ,in: Resources reserved for RSU use and Resources used by vehicles on the left and right roads respectively; To reduce potential conflicts caused by changes in the relative positions of vehicles traveling in the same direction, time slot resource sets... and All were further divided into , and ,in: For use by the lead car in the chain. For use by the last car in the chain. It is used by intermediate vehicles and isolated vehicles within the supply chain.

7. The method according to claim 4, characterized in that, The aforementioned collision-free slot allocation refers to eliminating access and merging collisions during packet transmission by using disjoint slot sets and dynamic intra-chain role changes. (Vehicle node) Not only does it broadcast its own driving status and decisions, but it also shares the time slot usage information of vehicles in the OHS set and the chain structure they belong to. Vehicles on the road should first listen to the channel to determine the time slots occupied by their THS nodes and the nodes in their chain structure, and determine the set of unoccupied time slots. The vehicle randomly selects a time slot from the intersection of the available time slot set and its corresponding set to access the network. When an isolated vehicle in one lane encounters another isolated vehicle, the probability of them choosing the same time slot is finite, since they both randomly select time slots. The available time slot, which is in the first... The probability of using the same time slot in consecutive frames is .

8. The method according to claim 4, characterized in that, The aforementioned collision-free time slot allocation includes: a) Scenario on the same lane: before the chain structure completes its growth. Always the lead car, Always acting as the tail train, since the time slot sets of different identities do not intersect, the two must use different time slots. Therefore, no transmission collision will occur during the growth process, and during the approach process, from The system obtains the slot allocation for the slower chain ahead and adds this information as control information to the broadcast data packet. Other vehicles in the faster chain adjust their broadcast slots according to the platoon slot information occupancy. The other three growth scenarios and disconnection scenarios are also handled in a similar process. b) Scenario on different lanes: The faster chain i is in a different lane from another chain. When they meet, their respective chain structures do not change. Received chain The last car When the beacon is sent, it understands the chain. All members and their time slots are analogous to chain merging on the same lane. All members in coordination chain i, select chain If a member does not occupy a time slot, select another unoccupied time slot to avoid conflict with the chain. In A conflict occurred; c) Intersection scenario: When an isolated vehicle or the leading vehicle stops, it should turn off its radio and notify the vehicles behind it, even if it does not broadcast, the following vehicles will calculate... When there is still a stationary vehicle, and the green light is on for the north-south direction and the red light is on for the east-west direction, if the behavior of the straight-going vehicles is the same as that of the non-intersection scenario in different lanes, the stationary vehicles in lanes 1 and 5 turn off their radios to avoid interfering with the broadcasts of vehicles traveling north-south. If vehicles in the right-turn lanes of lanes 1 and 5 want to turn right at this time, they will join the main road traffic as isolated vehicles, that is, the situation will change to an isolated vehicle encountering a chain structure or an isolated vehicle encountering another isolated vehicle.

Images