This series introduces the solutions, commercial value and typical cases of 5G Internet of Vehicles empowering self-driving sanitation vehicles, mainline logistics, terminal logistics, mining trucks, port automatic driving, automatic shuttle vehicles, Robotaxi, buses, etc. I have previously introduced 5G Internet of Vehicles empowering self-driving sanitation vehicles , 5G Internet of Vehicles empowering trunk logistics , 5G Internet of Vehicles empowering terminal logistics , and 5G Internet of Vehicles empowering mining trucks . This article introduces 5G Internet of Vehicles empowering automatic driving in ports . Please look forward to the next article, 5G Internet of Vehicles empowers automatic shuttle bus.
Text | Wu Dongsheng
The full text is 6,000 words, and it is expected to read for 16 minutes
(one)
Overview of the Port Autonomous Driving Industry
As a hub of transportation, ports play a pivotal role in promoting international trade and regional development. About 90% of global trade is carried by the shipping industry, and operational efficiency is crucial to ports. In 2020, China's major ports will complete a cargo throughput of 14.55 billion tons and a container throughput of 264.3 million TEUs (Twenty-feet Equivalent Unit, TEU).
In November 2019, nine departments including the Ministry of Transport issued the "Guiding Opinions on Building a World-Class Port", proposing the construction of an intelligent port system. Strengthen independent innovation and integrated innovation, increase the research and development, promotion and application of key technologies for port machinery and other equipment, automated container terminal operating systems, and remote operation control technologies, and actively promote the construction and transformation of a new generation of automated terminals and storage yards. Build an information infrastructure based on 5G, Beidou, Internet of Things and other technologies, promote the demonstration of automatic driving of collection trucks in the port area and collection and distribution channels in special scenarios, and deepen the linkage between the port area. By 2025, some coastal container hub ports will initially form an intelligent system with comprehensive perception, ubiquitous interconnection, and port-vehicle collaboration. By 2035, the container hub port will basically have an intelligent system.
The container terminal adopts the gate-yard bridge-container truck-quay bridge system, and cooperates with the loading and unloading equipment for terminal operations, yard hoisting and horizontal handling equipment, as shown in Figure 1. The loading and unloading process of containers at the port usually involves three operations [1] .
①The goods pass through the quay crane equipment and are loaded and unloaded on the wharf.
②The goods are transported between the wharf surface and the storage yard through horizontal transport equipment.
③The goods pass through the bridge equipment and are loaded and unloaded at the yard.
The horizontal transportation of containers refers to ② the operation link, and the goods are transported between the dock surface and the yard according to the designated path.
Figure 1 The circulation process of containers in the port
Horizontal transport of containers in ports faces many challenges .
(1) Difficulty in recruitment
The weather environment in the port is harsh, drivers need to work in shifts 24 hours a day, the work intensity is high, the environment is difficult, and it is not attractive to young people. 51.5% of the drivers are over 35 years old, and there are not enough new drivers.
Port transport truck drivers must hold an A2 driver's license, and at least 6 years of driving experience (3 years of C1 and 3 years of B1/B2) are required to obtain an A2 driver's license. Only 48.5% of truck drivers hold A2 driver's licenses, resulting in a labor shortage.
(2) High management cost
In my country's container terminals with a throughput of more than 1 million TEUs, a total of 6,000 to 8,000 internal collection trucks are deployed, with about 15,000 to 20,000 internal collection truck drivers. At present, the cost (including wages and social insurance) of Neiji truck drivers averages about 150,000 to 200,000 yuan per year, and it increases year by year. In addition, the current port transportation mostly uses diesel trucks, which consumes a lot of energy, and the cost of manpower and energy consumption accounts for more than 65% of the total cost of the port.
Port automation solutions usually have the following three ways [2] .
(1) AGV program
The application of Automatic Guided Vehicle (AGV) in ports generally uses magnetic nail navigation technology, which requires a lot of magnetic nails to be pre-laid in the wharf construction. However, the magnetic nail navigation technology is sensitive to metals. Once there is too much metal near the magnetic nails, the AGV navigation will be greatly disturbed. It is required that conventional reinforced concrete structures cannot be used in the civil engineering of the entire terminal.
(2) Self-driving straddle carrier solution
Autonomous Straddle Carrier (ASC) has a high cost per vehicle and integrates horizontal and vertical transportation. It is suitable for yard applications with less than 5 layers of stacking, and is widely used in Europe, America, and Oceania. my country's container terminals generally have dense storage yards, and multiple layers of containers are stacked to ensure space utilization efficiency. In my country's container yards, tire-mounted and rail-mounted container gantry cranes are widely used.
(3) Automated driving collection truck solution
Compared with other port automatic driving solutions, the cost of self-driving collection trucks is lower, and it is limited to horizontal transportation applications, suitable for all types of yard applications, and is mostly used in Asian ports.
Under comprehensive consideration, self-driving collection trucks have low requirements for site transformation, low single-vehicle cost, and flexible and convenient use. They are the optimal solution for the automation transformation of horizontal transportation in old and new ports.
The technical difficulties in the implementation of autonomous driving technology in ports lie in the special operating environment, high requirements for operating precision and low degree of scene standardization .
(1) Special operating environment
Autonomous driving collection trucks need to continue to operate in extreme weather such as heavy rain, heavy fog, and typhoon, which has high requirements for autonomous driving perception and decision-making. At the same time, the high-salt and humid operating environment will also accelerate the wear and tear of hardware equipment and increase replacement costs.
(2) High operating precision requirements
Port metal containers and large-scale infrastructure equipment will interfere with the positioning signal of the self-driving collection truck and expand the high-precision positioning error of GPS-based RTK. When autonomous driving trucks cooperate with medium and large-scale mechanical interactive operations, they are integrated into the terminal production business process, and need to achieve centimeter-level alignment accuracy requirements. For example, when interacting with suspension bridges, the error of the stop distance should not exceed 5 cm.
(3) Low degree of scene standardization
Although the port is a semi-enclosed environment with low speed restrictions, container loading and unloading is flexible, container stacking forms and road trajectories are subject to frequent changes, and the environment is highly dynamic. For example, when the area of the storage yard is insufficient, or the number of temporary containers increases, the roads in the port area may be used for temporary stacking of containers.
(two)
5G Internet of Vehicles Empowers Autonomous Driving in Ports
5G+C-V2X can meet the data transmission requirements involved in port self-driving collection truck perception and decision-making sharing, remote control, smart road perception and decision-making sharing, video surveillance and AI recognition, and help port self-driving collection truck fleets in smart vehicle management System and TOS system management to achieve collaborative work [3] .
(1) Autonomous driving collection truck perception and decision-making sharing
5G+C-V2X can provide map dynamic location sharing and tracking, real-time path planning adjustment, real-time information collection of internal and external sensors, real-time calculation and fusion of internal and external data of vehicles, etc. for autonomous driving collection trucks. The data volume of self-driving collection trucks is large, up to 40TB for a single vehicle per day.
The self-driving collection truck can also exchange information such as position, direction angle, speed and acceleration with the vehicle directly in front through C-V2X direct communication in real time. When there is a risk of collision, the autonomous driving collection truck makes timely decisions based on the information of the vehicle ahead, avoiding collision accidents and improving the safety of autonomous driving.
(2) Remote control
The automatic driving collection truck will also have remote control capabilities. When the automatic driving collection truck fails in the workplace, the operator can view the surrounding environment through the camera, make fault judgments, and remotely operate the automatic collection truck to exit the fault area.
In addition, rail cranes and tire cranes are the two most widely used gantry cranes in container terminals. The rail crane moves on the rails in the yard; the tire crane is equipped with tires, which is flexible and can be used for transfer operations. At present, tire cranes are mostly used in existing terminals, and rail cranes are mostly used in newly built terminals, and tire cranes account for a high proportion of existing terminals. The height of the gantry crane is about 30 meters, and the driver's cab is on the top of the gantry crane. The driver's working conditions are difficult, and the on-site operation is prone to fatigue and potential safety hazards. In order to ensure 24-hour operation, each gantry crane is equipped with three drivers in rotation. A wharf usually needs hundreds of gantry crane drivers, which has a high demand for drivers. Gantry cranes are remotely controlled through 5G large-bandwidth and low-latency networks, which can greatly reduce labor costs.
The main business unit in the loading and unloading operation area is the bridge crane. The height of the bridge crane is 60-70 meters, and the height of the electrical room is 50 meters. Wireless network is required to achieve network coverage in the operation area. In the remote control scenario, the number of simultaneous return cameras of a single bridge crane and the resulting uplink bandwidth requirements are several times that of the gantry crane. At the same time, the deployment of bridge cranes is relatively dense. Usually, 8 to 12 bridge cranes will be deployed on a 1-kilometer-long port coastline. In addition, because the vertical and horizontal moving speeds of bridge cranes are higher than those of tire cranes, the remote control has higher requirements for time delay.
(3) Intelligent road perception and decision-making sharing
5G+C-V2X can provide real-time map positioning update, real-time route planning adjustment, real-time road information collection, real-time road data calculation and sharing, etc. for smart roads.
The roadside equipment uploads the perception information of road obstacles (such as temporary containers, etc.), road conditions (such as water, icing, etc.) and other road conditions through sensors to the cloud platform in real time through the 5G network, and the cloud platform conducts intelligent analysis. Then send the road event information to the possibly affected autonomous driving collection trucks in real time; or analyze the perception information locally through the edge computing device deployed on the roadside, and then the RSU sends the road information to the possibly affected autonomous driving collection trucks, thereby Avoid accidents.
(4) Video surveillance and AI recognition
Video surveillance is very common in ports, and many areas in the port area cannot deploy optical fibers. For temporary deployment scenarios and mobile scenarios, wireless backhaul, as a supplement to optical fibers, has the advantages of flexible deployment, convenient adjustment, and low cost. Specifically, it can be applied in: AI recognition of container code ID by crane camera, automatic tally; intelligent analysis of driver's facial expression and driving status, early warning of abnormal phenomena such as fatigue and drowsiness; license plate number recognition, face recognition, cargo recognition management ; Use drones and robots to quickly and intelligently inspect.
(5) Integrate into port TOS system
The automation of container terminals is a systematic project. After the terminal operating system (Terminal Operating System, TOS) issues instructions, it needs to rely on the equipment control system (Equipment Control System, ECS) to execute. The TOS system is responsible for the highest level logic control of the terminal, including ship planning, container inventory maintenance, work order generation, gate operation, etc. TOS is responsible for releasing the highest-level tasks, informing ECS of the start and end points of port operations, and then the operating routes and corresponding actions of mechanical equipment such as gates, quay cranes, and storage yards are driven and executed under the control of ECS.
The decision-making planning for vehicle driving is divided into global planning and local planning. Global planning refers to the vehicle's start/end position planning and specific route planning; local planning refers to how the vehicle chooses a specific lane during road driving. Among them, the start/end position in the global plan is automatically generated by the TOS system according to the operation logic.
The self-driving collection truck is integrated into the port TOS system through the fleet management system, and docked with the port to realize functions such as background unified dispatching, route planning, vehicle remote monitoring, and intelligent management.
(three)
The commercial value of autonomous driving in the port of the Internet
The upstream of the port autonomous driving truck industry chain includes suppliers of key components such as sensing equipment (cameras, millimeter-wave radar, lidar, etc.), computing platforms, high-precision maps and positioning, and wire-controlled chassis.
The middle reaches of the port autonomous driving truck industry chain are solution providers and OEMs/mechanical equipment manufacturers.
The downstream of the port self-driving truck industry chain is the port enterprise.
There are two main business models for autonomous driving.
The first is the asset-light model . Autonomous driving technology companies and OEMs/mechanical equipment manufacturers provide port companies with large-scale mass-produced pre-installed self-driving trucks, or carry out post-installation fusion transformation. In addition, the overall solution for automatic driving in ports needs to be integrated with the TOS system in addition to the automatic driving collection truck. Therefore, the automatic driving technology company will also provide various technical services to port enterprises. as shown in picture 2.
Figure 2 Asset-light mode of port autonomous trucks
The second is the asset-heavy model . Self-driving technology companies purchase customized self-driving trucks from OEMs/mechanical equipment manufacturers, and self-built or joint-venture self-driving engineering contractors are established by self-driving technology companies to provide container horizontal transportation services to end customers. This new model of commercialized "on behalf of operation" is actually charged according to the volume of the container, which greatly reduces the purchase funds of the final customer in the early stage. As shown in Figure 3.
Figure 3 Heavy asset model of port autonomous driving trucks
Since the speed requirements of the internal collection card are not high, it is generally equipped with a 16-line laser radar, coupled with a perception system based on millimeter-wave radar and camera. The cost of a self-driving collection truck is about 1.2 million yuan. A typical internal collection truck needs to be equipped with 4 drivers "three shifts", each driver's annual salary is 150,000 yuan, and the total labor cost is 600,000 yuan per car. After adopting the self-driving technology, the cost can be recovered in about one year, and some safety hazards of manned driving can be eliminated at the same time. As shown in Table 1.
Table 1 Profit changes brought about by self-driving trucks in ports
Traditional port collection card |
Port automatic driving collection card |
cost comparison |
|
Purchase cost / 10,000 yuan |
50 |
120 |
|
Labor cost / (10,000 yuan / year) |
60 |
0 |
|
Purchase and labor costs after one year of use / 10,000 yuan |
110 |
120 |
-10 |
Purchase and labor costs after 2 years of use / 10,000 yuan |
170 |
120 |
+50 |
Purchase and labor costs after 3 years of use / 10,000 yuan |
230 |
120 |
+110 |
5G+C-V2X shares the real-time road information sensed by intelligent roadside technology with the port's automatic driving collection truck, so that the vehicle has the ability to perceive beyond the line of sight, predict danger in advance, change the path planning, and reduce the number of single-vehicle sensors.
It will improve safety, make up for the blind spot of vehicle perception, expand the range and distance of perception, and effectively avoid the impact of various obstacles in the port, especially the random container obstacles; it can reduce the number of sensors on a single vehicle and the requirements for the computing power of a single vehicle, effectively saving Bicycle cost: Perceive road conditions in advance, plan routes, reduce accidents, and effectively improve bicycle work efficiency.
(Four)
Typical Cases of Autonomous Driving in Networked Ports
Case 1: Xijing Technology's full-time unmanned electric truck Q-Truck
Q-Truck completely cancels the human cab, and installs a battery and cooling system on the front of the car to protect the core hardware system from complex working conditions such as port operations. The vehicle is equipped with industrial-grade sensors such as binocular artificial intelligence cameras, laser radars, and millimeter-wave radars, and cooperates with a complete set of full-stack systems to output ultra-high-precision positioning and identification functions, so that Q-Truck does not need to lay magnetic nails underground for guidance. It can guarantee a battery life of 200 kilometers and a load of 80 tons. It can successfully complete intelligent operations such as vehicle turning and bypassing, safe overtaking, etc., and realizes deep coupling of port scenes. For example, under the quay bridge and yard bridge, the alignment error between the spreader lock and the container lock hole does not exceed 3 cm, and the accuracy is close to 100%.
In the port area, the C-V2X global perception vehicle-road coordination system can realize the rapid cloud access capability of the unmanned mobile terminal, solve the problem of Q-Truck’s mobile network stability in the unmanned scene of the terminal, and ensure the internal realization of the terminal Pure remote automatic driving, stable interconnection of vehicle-to-vehicle, vehicle-to-dock control center network.
The fleet management system FMS (Fleet Management System) independently developed by Xijing Technology can effectively manage the fleet in the port scene, perform one-to-one task matching, reduce the empty driving rate of vehicles, monitor real-time positioning and control vehicles and goods Condition.
The fleet system can manage multiple unmanned trucks at the same time, and even realize mixed operation between unmanned and manned vehicles. In the event of unusual traffic conditions or bad weather, the Q-Truck remote control system can also realize manual takeover, and remotely control the vehicle such as emergency parking through the on-site communication system.
For different application scenarios, Xijing Technology has also developed the Q-simulation scene rapid simulation system to quickly build a simulation platform to simulate the use status for the user, and use this to design the user's scene [4 ] .
Case 2: Mainline Science and Technology Port Driverless Trucks
Facing the different market demands of traditional wharf renovation and upgrading and new smart container wharves, Mainline Technology has successively launched two port unmanned collection trucks. An electric pickup truck designed in a conventional shape can realize man-machine co-driving. In addition to enabling fully unmanned driving functions in ports, it also meets the needs of customers for "multi-scenario use" at the current stage. The other is the ART artificial intelligence transportation robot, a new generation of unmanned electric trucks specially developed and designed for new intelligent container terminals. It completely removes the cockpit design in terms of appearance. It is lighter in weight, lower in height, and narrower in width. It can drive in both directions in both directions. It has both completely unmanned driving and 5G remote driving functions, and can perfectly adapt to unlocking on the ground on the horizontal shoreline. craft.
In January 2020, Mainline Technology realized the commercial delivery and formation operation of 25 unmanned electric trucks in Tianjin Port for the first time. 25 unmanned electric trucks have completed 130 real ship operations in the Tianjin Port Autonomous Driving Demonstration Zone. The operating efficiency has continued to improve, reaching 33 natural containers per hour, the energy consumption of a single container has dropped by 20%, and the comprehensive operating cost has dropped by 10% %.
In October 2021, Mainline Technology assisted Tianjin Port to officially complete the full-scale operation of the world's first full-process automation upgrade project for traditional container terminals, and successively delivered 6 unmanned electric trucks to the intelligent container terminal in Section C of Beijiang Port Area With 60 ARTs, the world's first smart zero-carbon terminal was put into operation.
ART is equipped with "Trunk Master", an L4-level automatic driving system developed by Mainline Technology, which achieves high-precision, fully unmanned, all-weather, safe and stable container autonomous horizontal transportation operations in real complex operation scenarios in ports, effectively improving the overall port performance. Operational efficiency. At the same time, ART adopts new energy power, which can realize functions such as autonomous charging and intelligent algorithm power saving, reduce energy consumption, reduce operating costs, and truly achieve "zero carbon" emissions [5] .
references
[1] Chentao Capital. Autonomous Driving Empowers Smart Ports [N]. 2020,10.
[2] Yiou Think Tank. China's High-level Autonomous Driving Port Application [R]. 2020,10.
[3] Huawei, Shanghai Zhenhua Heavy Industry, China Mobile, Vodafone. 5G Smart Port White Paper [R]. 2019
[4] Yiou. Defining the commercial value of port digital intelligence transformation with real unmanned trucks [N]. 2021, July.
[5] Mainline Technology. Commercial delivery breaks 100, Mainline Technology lands the world's largest fleet of self-driving trucks [N]. 2021,11.
- END -
▎Recommended reading
▎Good book recommendation
01
With the acceleration of the industrialization of 5G Internet of Vehicles, following the publication of "5G and Internet of Vehicles Technology" in 2020 and "From Cloud to Edge: Edge Computing Industry Chain and Industry Applications" in 2021, in 2022 Dr. Wu Dongsheng will lead the "5G Industry Application "The author team launched another masterpiece - "The Future of the Internet of Vehicles: 5G Internet of Vehicles Innovative Business Model". Published by Chemical Industry Press, this book focuses on the business model of 5G Internet of Vehicles and explores the future sustainable development of Internet of Vehicles.
Dr. Wu Dongsheng Editor-in-Chief
Wu Dongsheng, Ph.D., Southeast University. He is currently the senior vice president of Gosuncn Technology Group Co., Ltd., the vice chairman of the Guangdong-Hong Kong-Macao Greater Bay Area Autonomous Driving Industry Alliance, the director of the Guangzhou Vehicle-Road Collaborative Industry Innovation Alliance, and the director of the Operation Center of the Guangzhou Intelligent Networked Vehicle Demonstration Zone. Committed to the research and application innovation of 5G, intelligent network connection, automatic driving, big data, artificial intelligence and other technologies. Published dozens of papers in provincial and municipal periodicals, edited books such as "5G and Internet of Vehicles Technology", participated in the compilation of "Guangzhou Intelligent Networked Vehicle and Smart Transportation Industry Development Report (2020)", etc.
purchase link
02
Published by People's Posts and Telecommunications Press, this book focuses on 5G and interprets edge computing from a point-to-plane perspective . On the basis of introducing the connotation and core technology of edge computing in the 5G era, it focuses on analyzing and introducing the situation of the edge computing industry chain and the application of seven typical industries of edge computing. The industrial chain covers upstream, midstream and downstream. Typical industry applications include the transportation industry (autonomous driving, intelligent network connection, intelligent transportation, and smart roads), security industry, cloud gaming industry, industrial Internet, energy Internet, smart city, and smart home.
purchase link
03
This book systematically analyzes the overall architecture, system standards, key technologies, typical services and application scenarios of 5G IoV, comprehensively scans and analyzes the latest developments in global and domestic IoV, and analyzes the challenges and challenges faced by the 5G IoV industry development. Prospecting and forecasting future development prospects is of reference value for practitioners who are committed to 5G and Internet of Vehicles industry research, standardization and related product realization.
purchase link
about Us
"5G Industry Application" is a research and consulting platform that gathers senior experts in the TMT industry. It is committed to providing enterprises and individuals with objective, in-depth and highly commercially valuable market research and consulting services in the 5G era, helping enterprises to use 5G to achieve strategic transformation and business refactor. This official account focuses on providing the latest developments and in-depth analysis of the 5G industry, covering communications, media, finance, automobiles, transportation, industry and other fields.
