In the face of these rising demands, the transport industry is now seeking solutions that will take existing transport infrastructure to the next level. Digitalization not only plays a key role for mobility providers in their struggle to achieve the reliability, flexibility and availability they need to make rail solutions cost-efficient: It lies at the heart of making the lives of people who travel easier and more enjoyable.
Locomotives, trains, components, train automation systems — everything is a valuable source of data. With Siemens Digital Services, a state-of-the-art system that processes this wealth of transportation data, transport operators analyze trends and monitor the status of machinery and the flow of passengers and goods to offer a smooth mobility experience.
The Turkish high-speed rail service YHT takes great pride in a level of hospitality that turns passengers into guests who surf the web, enjoy films and play — right there in their seat. In the past, reaching the other side of the Alps has been an adventure. Today the journey through the Gotthard Base Tunnel is safe and comfortable - thanks to the Siemens solutions for tunnel safety and fire protection. The ICE 4 sets new standards for rail travel in Germany. Learn more. Siemens initiated the creation of an art based visualization as an ideal way to make data more tangible as well as create a unique feeling and understanding of how data works.
And, last but not least, digital solutions ensure an enhanced and contemporary travel experience with constant Internet access and customized services. Are you interested to know how digitalization today is being applied across the mobility sector and how data can drive value for society, operators, and authorities? Reduced weight and optimized aerodynamic design reduce energy consumption per seat by 22 percent compared with an ICE 1. The ICE 4 is a true world champion when it comes to usable floor space, passenger in a meters train. Advanced traffic management system integrating transport information for all travel options.
To deepen the vehicle-specific knowledge in the study of automotive engineering are used modules such as vehicle development, design, drive and suspension technology, mechatronics, vehicle dynamics, control and driver assistance systems, measurement and testing technology, and CAD simulation. Depending on the range of courses, specialization is possible in areas such as design and development, system and drive technology, energy technology, suspension or bodywork technology, service or design. In addition, there are modules such as cost and investment management, business organization and industrial management, project and quality management and English.
Degree programs in the field of electro mobility impart scientific and practical basic knowledge in mathematics, physics, electrical, information and materials engineering as well as in mechanics. The Automotive Engineering degree course focuses on the areas of mathematics, computer science, physics, chemistry, material science and engineering.
Subsequently, the knowledge in the various automotive application areas is deepened and expanded. Supplementary voting modules are e. Logistics, prototype production, industrial accounting or technical English. Based on engineering basic subjects, the traffic engineering knowledge in areas such as planning and operation in transport, mobility and traffic research, traffic management and telematics, vehicle technology, testing and approval and, if necessary, aerospace engineering or ship and ocean technology will be expanded.
In addition, there are modules from the economic, legal, humanities and social sciences. The course Aerospace Engineering teaches economic, legal and social science knowledge in addition to the engineering and technical fundamentals. There are also application-related modules such as aerodynamics, design and construction, manufacturing technology, flight mechanics, lightweight construction, engine construction, machine dynamics, control engineering, fluid mechanics and materials technology.
In addition,Technical English, business and work organization, cost accounting or management knowledge. In rare cases, the study is combined with a pilot training ATPL. Depending on the offer, the master's program focuses on areas such as aircraft construction, lightweight construction technology, engine or space technology. In the study of marine and marine engineering , subjects such as welding technology, hydromechanics, ship elements, buoyancy and stability are on the curriculum in addition to the engineering fundamentals.
This is followed by special application modules such as shipbuilding, ship equipment, design and construction of special cargo vessels, yacht design, underwater vehicles, marine engineering, ship electronics, shipbuilding, manufacturing and shipyard operations. In addition, students receive interdisciplinary training in the areas of business administration, project management, quality assurance and law. They take part in practical work study programs in the form of industrial or business projects.
A specialization is partly possible in subjects such as ship design, hydrodynamics and yacht design, construction, ship safety and marine equipment. Agents work in a perceive—reflect—act cycle that defines its dynamic behavior. This cycle is modeled with three types of components i. The sense meta-class models perception as an interaction with observed devices that adds or modifies facts in the state of the agent or the sense. The instances of the reflection meta-class are responsible of the reflection stage. They analyze available knowledge both facts and goals , and produce new one or modify the existing.
Then, they check non-satisfied goals , and identify the tasks related to them and suitable for execution in the current context. Finally, the acting meta-class represents the actual execution of one of the executable tasks. The agent concept is suitable to represent directly control managers. For persons involved in traffic, there are some modifications.
A person has specific person sense and person acting components that allow interacting directly with spots without the need of intermediate devices. The person sub-type of agent is also a subtype of the TML person. Regarding the environment , it represents the elements participating in traffic that are not people, their vehicles, or the ITS components e. Examples of its elements are the roads, the traffic signals, the weather conditions, or obstacles. It is an extension of the TML environment see Figure 6. It is modeled as an aggregation of things , which are located over a map. The map represents the distribution of the physical space.
It is a kind of graph composed by a set of section , each one linking two junctions at least one and having a given length. This length can be measured in different units, e. A section is a set of lanes that can be direct from the origin to the destination junction or reverse the other way. A lane has an occupants list of references to the vehicles or persons in it ordered according to their position. They can be placed in specific map locations , but also being considered global for the environment, e.
Vehicles and persons , being places , can have a location. In the case of instances of vehicle , this is mandatory. Persons can have a location, when they are pedestrians, or being attached to a vehicle. The ITS metamodel also includes general instantiation, inheritance, and association mechanisms. For each meta-class of name name , there is another meta-class of name Iname that represents its instances at the model level. An instance of the later can be linked to an instance of the former with an instance of the meta-relationship instanceOf.
For example, there is a meta-class IPerson linked by a meta-relationship instanceOf to the meta-class Person , so in a model, a specific driver IPerson can be defined as an instance of a specific type of Person. It can be applied to most of concepts in the ITS metamodel, with the semantics that the sub-concept inheriting from the super-concept has all its features, but it can extend or constrain them with additional elements.
Default constraints establish that an animated concept i. Regarding association, there is a general meta-relationship relatedTo between instances of component. It has a stereotype property to establish arbitrary types of it. Figure 7 shows the development guideline for the framework. The process has two different stages.
The definition of its metamodel see Section 3. The results of this stage are models that provide an abstract representation of the ITS and its environment. They are independent of specific target simulation platforms. Concepts starting with uppercase belong to this ML. The process starts identifying places in decision 1.
These establish the context of the system and the elements over which the ITS must work. If there is any place , activity 2 assigns it a name and identifies its spots. Next activities in this path specify the place as a component starting with decision 3 and its spots see decision The description of a place depends on its type see decision 3 : thing , vehicle , or person. All of them have to be specified as components see activities 4—6. This specification indicates the information available from these elements, and the means to interact directly with them.
- How to Play Poker: Tips and Strategies for Beginners (Revised Edition)!
- Intelligent Freight Transportation!
- Mud On The Tires.
- View Intelligent Freight Transportation Automation And Control Engineering 2008;
Instances of person should not usually include any external method to modify its internal knowledge, as it would violate their autonomy according to the agent paradigm. Other of their features can be modified from the environment, for instance to reflect that they have suffered some injury. If needed, these places must also be located in the environment map see activity 13 , which also implies specifying this.
The specification of a person also comprehends its behavioral cycle. This is specified in terms of its sense see activity 10 , reflection see activity 11 , and acting see activity 12 components. Sense instances use their methods to get notifications from external sources i.
Reflection instances model several activities: generation of new derived knowledge from that already available; checking goal satisfaction according to the current internal state; selection of potential tasks to execute according to the state of goals and other knowledge.
Acting instances represent the actual execution of tasks. Activity 7 is used to specify those goals , activity 8 the facts , and activity 9 the tasks. These elements are linked among them: tasks are potentially able to satisfy some goals ; goals are satisfied according to conditions specified in terms of knowledge ; and tasks are eligible depending on the state of associated goals. After specifying a place , the process continues with its spots.
These spots are elements where devices can be located, or that devices can observe and act on. The specification of spots starts identifying them see decision Every spot must be defined as a component to determine its related information and methods see activity Spots are also the elements that can have attached containers see decision Activity 17 identifies the containers attached to a spot and the devices running in it.
Containers are connected through the channels identified in activity Among devices , sensors observe spots and actuators act on them. These spots can be or not the same to which their containers are attached. Activity 18 links sensors and actuators to their target spot. Finally, activity 20 specifies the container devices as components and their managers as agents , if they exist.
Although not included in the figure, the specification of managers should follow a similar workflow to that of person see activities 6—12 , as both entities are sub-types of agent. The main issue here has to deal with code generation. It relies on platform-specific templates see Section 2. This approach is in line with MDA [ 26 ]. Figure 7 shows only a sequential perspective of the process. In fact, its actual application needs to be iterative in its different paths. For instance, in the path from nodes 3 to 13, activities 4—6 can be done in the first iterations to give a high-level description of the system context, while the other activities correspond to a later and more detailed specification.
In that path also, when modeling the person tasks see activity 9 , it is possible to find out that their execution requires additional internal methods of person , so activity 6 must be revisited. Moreover, when in activity 18 designers link sensors and actuators to the spots they work on, this can lead to modifications in the definition of those spots made at some moment in activity 15 , or even of their places starting in decision 3.
The previous process is supported by two main tools. A graphical model editor allows generating model instances compliant with the ITS metamodel.
Vehicle and traffic engineering
A code generator takes as input the metamodel and the model instances to generate the source code of the simulation for a target platform. Researchers only need to provide the code templates for that platform, and give the correspondences between those templates and the concepts of the ITS metamodel. The code generator has been developed for the traffic simulation framework [ 12 ]. This section illustrates the application of the previous framework see Section 3. It includes a case study see Section 4. The availability of affordable sensors and actuators for ITSs has made traffic experts reconsider some of the traditional designs of traffic facilities.
In particular, traffic signals can be made more intelligent , in the sense of being aware of the state of traffic in their areas and coordinate with other signals to improve the overall state in wider areas. The use of intelligent traffic lights to control junctions in streets is part of this approach [ 16 ]. In that work, the intelligence comes from two main features of these signals.
First, they are able to perceive the number of cars in the lanes that join in their junctions. Second, they exchange information to adapt their behavior to the overall traffic state in the area with the goal of maximizing the throughput of cars from it. In order to evaluate the benefits of the use of these intelligent traffic lights over traditional ones, the work in [ 16 ] simulates both situations using the SUMO platform and real data of traffic. This case study reproduces this experiment using the ITS simulation framework.
The process starts with the identification of potential places in the problem see decision 1 and activity 2. According to its definition see Section 3. The presentation of the problem introduces several of these places. Junctions contain intelligent traffic lights where OIS sensors are located. These sensors perceive lanes from streets. At this point, the process can continue by specifying those places starting in decision 3 or the spots and related devices starting in decision 14 see Figure 7.
As the original work pays particular attention to the system devices , this case study continues in the second path. Figure 8 shows the model instance resulting of it. Decision 14 begins identifying the spots of the previous places. The traffic lights are considered as located in the junction posts surveying the main street sections.
Both junction posts and street sections are spots , as the former represent where devices are located, and the latter what they survey and affect. Activity 15 specifies these spots as components. In this case, they do not have any specific state or functionality to specify. Only street sections could have it related to the vehicles in them. However, this specification is considered later in the environment map. The next nodes in the path focus on the devices the simulated ITS uses to perceive and act. The first step in decision 16 is discovering the spots with containers for these devices.
In this case, there is only one of such spots : the junction posts. Activity 17 identifies the devices running in the containers. The intelligent traffic lights are modeled as containers in junction posts. They have as sensors the OIS sensors that allow them perceiving the queues in their lanes using the street sections ; their actuators are their red, green, and yellow lights. Activity 18 links OIS sensors to the street sections spots they watch, and lights to the street sections spots they are pointing at.
The containers of ITSs are not isolated. They set up a complete system thanks to its communications through channels. Activity 19 identifies these channels. The intelligent traffic lights are connected in this way not shown in the figure to share the state of the whole area. Activity 20 models the sensors and actuators of containers as components , and their managers. An OIS sensor has a method to consult the number of vehicles in its lane.
A light has a state characterized by its current color, the time remaining in it, and the current default time assigned to each color; its methods allow checking the previous values and changing the periods assigned to each color and which to display. As in the SNML, complex control algorithms are modeled using the managers of devices , which are agents. In [ 16 ], traffic lights adapt the time assigned to each color according to the information from their sensors and from traffic lights in the other junctions. For this problem, a lights manager orchestrates work in each intelligent traffic lights container.
It obtains information from its OIS sensors and sends it to the managers in the other junctions using the channels. It also receives from the channels the updates from the other junction posts. Finally, with all of this information, it updates the lights state. Figure 8 also shows how to introduce instances of the types identified for a given problem. The instance lights 1 of the class lights runs in the instance intelligent traffic lights 1 of the container , and it governs the traffic of the instance street section 1 of the class street section.
After modeling spots , containers , and their devices , the specification needs to consider the persons and their vehicles, as well as the environment where all these elements interact. This corresponds in the process to the paths for places starting in decision 3 see Figure 7. In [ 16 ], only vehicles are modeled, as drivers are implicit in their control. Using data from real places, researchers determined the volume of traffic in the different sections of streets and the junction turning percentages.
In the ITSML, the vehicle is a passive component , and the decision-making is modeled with the driver person. Activity 5 models the vehicle component. In this case, it has no state. The environment map models its position see later activity Its methods should move the car, change the lane, and check its own position, that of other vehicles around it, and the color of the traffic lights in its street section.
The specification of person is more complex. It includes the elements representing its basic knowledge and actions see activities 6—9 , and the perceive—reflect—act cycle see activities 10— Activity 6 specifies the person as a component , mainly regarding the external methods intended to check its state. Here, no one is needed. Activities 7—9 model its knowledge and capabilities.
The goal of the driver is the path to follow in the street. It is characterized by the entry and exit points to the street, and the exit to take at each junction. The agent only needs to know the position of the vehicle to make decisions. This is a fact that directly translates the information provided by the vehicle methods. The tasks of the agent are those to check the environment and manoeuver the vehicle. In the first group are checking junction to know when to turn and in which direction , checking position to know if there is a vehicle in a given position , and checking lights to know if the vehicle can go into the junction given the traffic lights.
The second group includes moving forward , turn left or right and the number of exit , and stop the vehicle. Activities 10—12 use the previous elements to represent the acting cycle of the driver person. Activity 10 models sense. It only needs to perceive the vehicle spot to assert the facts regarding its current position, whether the surrounding positions are busy, and the color of traffic lights. Activity 11 models reflection.
It compares the current position with the next turn.
If the vehicle has not reached the turning position and lights allow it, it chooses the task to move forward; if it has reached the position, it chooses to turn when the target position is available; in other case, it waits. Activity 12 models acting , that translates one of the potential tasks for execution to person and vehicle methods. The last places to describe are things in the environment. Activity 4 specifies them as components. The already mentioned junctions and streets are things. Here, their specifications are empty because these places are only containers for spots.
The final step in this modeling is activity 13, where places are located in the environment map. In [ 16 ], there are three junctions across a street, and each junction connects two sections of the street and a transversal section from other streets. The thing street is located over the three junctions and two sections. These elements and the transversal sections have direct correspondences in the ITSML as parts of the map. The model also needs to specify the number and orientation of lanes in each section. Figure 9 shows the model resulting of these activities, omitting lanes of sections and the links between the street and junctions.
All the elements appear as instances i. For instance, the metamodel see Section 3. Section 1—2 and section transversal 1—3 are instances of a class section that is a direct instance without any modification of the meta-class section. The reason of this modeling is that specifying these instances only needs using existing elements, and not adding new state, methods, or relationships.
In the case of multiple inheritances, the base target meta-class is that with the closer semantics. In this case, INGENIAS models mainly add the description of the interactions among manager agents regarding the exchange of information from their intelligent traffic lights. The map of the environment is represented in SUMO as a network.
Mobility | Products & Services | Siemens
SUMO does not allow controlling the movements of its vehicles , as it provides only a limited set of implementations for them. Nevertheless, the path vehicles have to follow can be specified, and SUMO controls that traffic signals are observed. Regarding traffic lights, SUMO supports actuated traffic lights for the experiments described in [ 16 ].
They can change the time showing each color according to the number of vehicles in the edges they control. The proposed framework has been tested in several experiments with the purpose of assessing its components. These experiments considered different contexts in order to evaluate its adaptability to multiple needs regarding simulation. In particular, the ITSML has been used to specify different models for people and types of ITS, and the process and tools also for several target platforms.
Regarding decision-making see Section 3. The work in [ 32 ] offers an alternative perceive—reflect—act cycle. Its agents consider timed data, and combine for reflection a reactive model and a short-time planner. For instance, hard and soft goals [ 14 ] are sub-types of goal , and the short-time planner [ 32 ] of reflection. Relationships can be represented using instances of the relatedTo meta-relationship.
Nevertheless, this is a limited approach, as it does not allow introducing constraints. There is also work with other types of ITS. In [ 33 ], there are eye cats a type of spot deployed on road borders which are things. The eye cats hold magnetic trackers as sensors that perceive vehicles passing near them. Manager agents represent the control of these sensors and their communications. The framework has also been tested with different target simulation platforms. MATSim can also be extended with additional behavioral modules, as the platform in [ 35 ].
The latter distinguishes between vehicles and drivers, and supports complex agent-based models for people. Drivers have a basic behavior of path following that can incorporate maneuvers and general constraints on movement as goals. With the ITSML, those constraints are also modeled as goals , linked to tasks that represent maneuvers. Tasks are related to conditions that represent when those maneuvers can be considered for execution. In all the cases, the first tests with these platforms require developing specific code snippets to connect already available code that for abstract models with the platform.
This also happened in the experiment reported in the case study. These new fragments could later be reused, or adapted with less effort, when experiments continued, and integrated in the code generation tasks. In these cases, the process points out the ITSML primitives suitable to model the concepts from those models and platforms.
However, two aspects remain little described, the choice of relationships and the use of inheritance. Relationships are deduced from the context provided by concepts. The process focuses on the latter, and these determine the allowed relationships. There are no indications on how to choose when several relationships are available beyond their description in the ITSML.
There are also sometimes difficulties to design inheritance hierarchies, particularly to establish what concepts must be extended. This work is related to two main aspects of research in ITSs. It deals with the relevant concepts to model this kind of system, and the approaches followed in their analysis, development, and simulation. ITSs cover a wide variety of needs [ 3 , 5 , 6 , 36 ], e. However, their architectures and perspectives on the environment present a similar structure.
- Gedichte (German Edition).
- That Loving Touch (Mills & Boon Vintage Desire).
- Everyday Corruption and the State: Citizens and Public Officials in Africa.
- Leading change through intelligent automation solutions.
- Keeping His Way Pure;
- The Price of Paradise Part 2 (The Broken World).