BR102014021691A2 - standalone transport hardware with road vehicles - Google Patents

Abstract

hardware for autonomous transport with road vehicles” the present invention relates to an autonomous road transport system based on autonomous vehicles operating in a similar way to trains on railways. in this system, a common road vehicle, such as a car, bus or truck, can be properly equipped with sensors and computers to drive itself, guided by virtual tracks from one station to another. a virtual railway is built from a set of stations connected by virtual rails. Virtual tracks can be any type of technique or equipment capable of guiding the vehicle through the center of the road autonomously, such as advanced mapping and location techniques or magnets installed under the pavement. the stations serve as a starting and stopping point for vehicles.

Description

Field of the Invention The present invention relates to systems and methods for enabling a road transport system to operate in an autonomous mode. This invention is correlated with position control, travel and position of land vehicles. For example, autopilots. It is also correlated with measuring the distance traveled on land by vehicles, for example using odometers.

Background of the Invention (002) Currently, the worldwide road transport system uses a human driver to safely drive the vehicle from origin to destination. The major problem with using a human driver is that he is subject to failure for various reasons such as tiredness, distraction, weakened emotional state, recklessness, malpractice, etc.

When comparing road transport with others, we realize that the task of driving the vehicle requires a high and permanent degree of driver attention throughout the journey. First is that the driver has to manually keep the car in the center of the road with an error tolerance of a few centimeters. Second, why he must be permanently aware of other vehicles and obstacles so that he can take actions to avoid collisions. This permanent state of attention causes stress to the driver, which can lead to fatigue and consequently to accidents. On the other hand, the reality of air, sea and rail transport is quite different. In these modes of transport, drivers act as supervisors during travel, as automatic guidance and autopilot systems take control of the vehicle. As a result, the driver receives a much lower stress load, which contributes to the reduction of accident risks and contributes to the increase of their quality of life.

[004] The present invention is based on the application of new technologies such as laser sensors and computer vision as well as mapping and location techniques to enable road transport to be able to operate with automatic guidance and achieve all the advantages described above.

[005] The technology of autonomous vehicles is not that new. In 1999 the ARGO research project, from the University of Parma - IT, developed a camera-based autonomous steering system that was able to drive a 2000-kilometer adapted vehicle on Italian highways. This technology is constantly developing in several countries and the advances made in this area can be seen in the DARPA Grand Challenge and DARPA Urban Challenge robotic vehicle competitions, which represent a milestone in the evolution of this type of vehicle. The development of Laser Range Finder sensors, capable of measuring the distance of surrounding objects at a 360 ° angle, of increasingly powerful computers, global location systems, and inertial navigation sensors have greatly contributed to the development. of mobile robotics technology and autonomous vehicles.

[006] Projects such as the Google - US Autonomous Car, the Parma - IT University Project VisLab or the Berlin - ALE University Autonomous Project show the full potential of this technology, but are still operating as prototypes.

[007] On a commercial scale, several vehicle manufacturers such as Volvo, VW, BMW, Toyota, Mercedes already have some of their standard Driver Assistance System (Driver Assistance System) in their standard vehicles, which can take actions such as warning if The car is leaving the lane, automatically adjusting the cruising speed keeping the correct distance with the vehicle in front, warn about collision risk among other things.

Automatic car steering requires some type of sensor and / or method capable of locating the vehicle in relation to the path. Commercial aviation systems are already widely used in commercial aviation, mainly based on a global location system such as GPS, which has location accuracy of a few meters. This accuracy error does not affect the automatic steering of the aircraft since a lateral deviation of a few meters does not hinder the operation. If the GPS system is used in conjunction with the Wide Area Augmentation System (WAAS), the location accuracy is increased to the point that the approach to landing can also be done automatically. Like aviation, shipping also uses the benefits of the global location system, as a mistake of a few meters in a ship's ocean route can be tolerated. In the case of rail, automatic guidance is already a natural feature of the system.

[009] In the case of road transport, automatic guidance is much more critical, as the vehicle must remain centered on a road with a tolerance of a few centimeters. For this reason the location techniques used in air and sea transport do not apply. In addition, automatic guidance must be able to handle the presence of other obstacles and react quickly. The operating principle is based on sensors installed in the vehicle to detect its position in relation to the road a control motor coupled to the steering wheel to correct the steering. Related Art [010] US8078349 patent and title "Transitioning a mixed-mode vehicle to autonomous mode" shows a transition method for unattended operation for mixed, manual or autonomous vehicles, but does not describe how the vehicle once operates. that it goes into standalone mode.

[011] US7610146 and title “Vehicle position determining system and method” shows location systems and methods for autonomous vehicles operating on highways, but is limited to the question of location and automatic guidance along the highway. The description of vehicle behavior in the context of a complete transport operation is not given.

[012] US Patent No. 4361202, entitled "Automated road transportation system", shows an autonomous road transport system. Despite the great reliability, the system has the disadvantage that it requires the installation of transponders for highway identification and the obligation of the metallic guide rail system.

SUMMARY OF THE INVENTION The applicant has found that the application of modern mobile robotics techniques such as mapping, locating, guide rails, sensors and obstacle detection methods can be inexpensively adapted to conventional road vehicles to operate as standalone vehicles. In this new model, the transport operation resembles the rail system, where the vehicle starts a journey at one station, travels guided by a rail system, in this case virtual, and ends the trip at another station.

[014] The advantages of this system are: • The driver goes on to work as a travel supervisor which improves his working condition and especially increases the safety of the trip as most accidents are caused by human error. • The proposed system can be adapted to any type of road vehicle such as bus, truck or car. Under well-signposted roads low cost sensor configuration is possible. • The proposed system may use different techniques and methods of automatic guidance. Under well-signposted roads it is possible to use virtual track techniques that do not require the installation of any special equipment or flags along the highway, dramatically reducing the cost of system assembly and maintenance.

Brief Description of the Figures [015] Figure 1 shows an example of an assembly of an autonomous virtual rail transport system.

[016] Figure 2 shows an example of mounting equipment on a vehicle so that it is able to operate autonomously on the virtual railway.

Figures 3A, 3B and 3C show examples of station assembly.

[018] Figures 4A and 4B show examples of mounting simple stop marks that can be used to define the exact stopping place of vehicles at stations.

Figures 4C and 4D show examples of mounting initial location marks that can be used to calculate initial vehicle positioning within the station.

[020] Figure 5 shows an example of setting up a virtual track system based on detailed road maps and precise location techniques.

[021] Figures 6A and 6B show an example of mounting a virtual rail system based on magnetic markers installed on the roadway.

Figures 6C and 6D show an example of mounting a virtual rail system based on a painted strip on the floor.

[023] Figure 7 shows an example of mounting a sensor system for obstacle detection.

[024] Figure 8 shows an example of an autonomous transport operation via a virtual railway.

[9] Figure 9 shows an example of a procedure that enables the vehicle to operate in autonomous mode.

[026] Figure 10A shows an example of a simple autonomous transport mission.

[027] Figure 10B shows an example of a two-step autonomous transport mission.

[11] Figure 11 serves as a reference for describing the mission examples of figures 10A and 10B.

Detailed Description of the Invention The examples of assemblies and methods given in the detailed description, figures and claims should not be construed as limiting factors. Other assemblies may be used, and changes may be made, without departing from the subject matter set forth herein. It will be readily understood that aspects of the present disclosure, generally described herein, and illustrated in the figures, may be arranged, substituted, combined, separated, and designed in a wide variety of different configurations, all of which are explicitly contemplated herein.

[030] The methods and systems disclosed herein refer to an autonomous road transport system based on autonomous vehicles operating similarly to railway trains. In the transportation model presented here, a common road vehicle, such as a car, bus, or truck, can be properly equipped with sensors and computers to drive itself, guided by virtual tracks from one station to another. In an example assembly, a virtual railroad is constructed from a set of stations interconnected by virtual rails. Virtual rails can be any type of technique or equipment capable of autonomously guiding the vehicle through the center of the highway, such as advanced mapping and location techniques or magnets installed under the pavement. 1. System for autonomous transport by virtual railways [031] Figure 1 shows an example of an assembly of an autonomous transport system by virtual railways. In this assembly example, the autonomous transport system (100) may have a autonomous vehicle (102), virtual rails (101), and stations (103). The standalone vehicle (200) may be equipped with a global location sensor (202) used to identify which station the vehicle is at, a home location sensor (203) for accurate determination of the vehicle position within the station, a virtual track tracking (204) used to keep the vehicle centered on the tracks during unattended operation, an obstacle sensor (206) to detect other vehicles and / or obstacles, a central computing system (201) for processing the information received sensors and an operator interface (206) through which a human operator can control the vehicle. In addition the vehicle can be further equipped with a dead reckoning unit (207) to measure incremental displacements. The virtual rails 101 describe the exact path the vehicle 102 should go through and also mark the exact location of the stopping points (stopping marks 104) within the stations. Each route of the virtual track can be traveled at different speeds according to traffic regulations. In an example station assembly (103), we have a station with a stopping point, defined by the stopping mark (104) and a start location mark (105) that is detected by the vehicle start location sensor (203). , to determine the exact position of the vehicle within the station.

[032] The two main functions of a station within the autonomous virtual rail transport system are: (a) To serve as a vehicle entry and exit point for the virtual rail system, in a process referred to herein as connecting or disconnecting with the virtual rail. . This process, explained in more detail below (p. 19-1), basically involves determining the starting position with great precision, (b) Serving as a stopping point for loading or unloading products or people during stand-alone operations.

[033] Depending on the operating needs of the autonomous transport system (100), some stations may operate only on one side of the highway (300) and others may operate on both sides (350). Each station may contain one or more stop markings (302) which may be, for example, any type of mark printed on the floor that delimits the space occupied by the vehicle when it is parked there. Also at each station there may be one or more starting location marks (303) which may be, for example, an appropriate graphic pattern on a board or even on the floor that is within the line of sight of the vehicle starting location sensor (203). ).

[034] The virtual track tracking sensor (204) that equips the vehicle may be, for example, a camera that uses available visual information along the highway, such as signs, ground markings or trees, to constantly calculate its position relative to the virtual track (101) described on a map. In another mounting example the vehicle may be equipped with more than one tracking sensor such as two cameras or a camera and a LASER scan sensor to further improve location accuracy and therefore the ability to track the rails. The vehicle may also have one or more obstacle sensors (205), such as cameras and / or LASER scan sensors to detect obstacles and other vehicles in the vicinity. In another example of mounting the start location sensor 203, the track tracking sensor 204 and the obstacle sensor 205 may be one with multiple functions. The global location sensor (202) allows the vehicle to identify which station it is in, but may not have sufficiently high accuracy to determine the vehicle's position relative to the virtual tracks by a few centimeters. In these cases it is necessary to use additional sensors.

[035] The main function of the virtual trail is to guide the autonomous vehicle in the routes between stations. It defines the exact path the vehicle can take when it is operating in autonomous mode and should preferably be located in the center of the highway. For multi-lane highways, there may be one track for each lane. When there is a fork in the road, the virtual trail can be split in two, where each trail continues along a path. In this case, the decision of which path the vehicle will take may be defined, for example, in the transport mission which vehicle is performing. The virtual rail must be detected by the virtual rail sensor (204) so that the autonomous vehicle (200) can follow it.

[036] An autonomous transport operation can begin when a driver drives the vehicle to a station (103), parks at a stopping point indicated by a stopping mark (104) and selects through the user interface (206) that the vehicle connect to the virtual railway (101). Once this procedure has been successfully completed, the vehicle is enabled to operate in autonomous mode and awaits the user's permission to begin the execution of an autonomous transport mission. An autonomous transport mission can be, for example, departing from one station and driving with controlled speed to another station. Arriving at the destination, the vehicle can, for example, stand still waiting for a new mission or be removed from the system by disconnecting from the railroad. In another mission example, the vehicle may travel through a given sequence of stations, stop at each station, and wait for permission to continue, which may be given by a human operator. Mission programming can be done on any computer, including the vehicle itself. If an external computer is used, scheduled missions must be copied to central computer system memory (201) for it to perform the mission.

[037] To connect to tracks, the vehicle can use the global location sensor (202) to identify which station it is in, and then use the home location sensor (203) to accurately determine its position and orientation within and, therefore, in relation to the virtual rails (101). In one embodiment example, the global location sensor (202) measures the overall position of the vehicle and uses this information in the vehicle's central computing system (201) database to find out which station it is in. Since the distances between stations are large compared to the global location sensor error, usually a few meters, there is no risk of erring in which station it is. In an example assembly, the home location sensor 203 may be a camera that "sees" the home location mark (105) of the station and uses this information to accurately calculate the position and orientation of the vehicle. To properly connect to the virtual rails (101) the starting position accuracy (x, y) must be a few centimeters and the orientation accuracy (Θ) in the order of tenths of a degree. Normally using only the global location sensor and compass this level of accuracy cannot be achieved. 2. Examples of station mounts.

Figures 3A, 3B and 3C show examples of stations that can be used in the autonomous transport system. Figure 3A shows a station assembly operating on one side of the road only, which has a stopping mark (302) for the vehicle to park and a starting location mark (303) for the vehicle to accurately calculate its position relative to the tracks. virtual resources (101). Fig. 3B shows an example of a station (320) that can operate up to three vehicles simultaneously. Finally, Figure 3C shows a station (330) operating on both sides of the highway, has two stop marks (302) and two start location marks (303). Other examples of mounting a station can be obtained by combining the characteristics of stations 300, 320 and 330.

[039] The stop mark (302) can be any graphic object that serves to indicate the location within the station where the vehicle should stop. Figure 4A shows an example of a stop mark that can be used to operate vehicles of different lengths such as cars and trucks. Dotted strips 401 are for side alignment and solid stripe for front alignment. Figure 4B shows an example of a double width stop mark 410 suitable for operating with vehicles of different widths. The stop mark also serves, for example, to show a human driver where to park the vehicle before beginning the precise location procedure preceding a transport mission. Another utility of the stopping mark is to indicate to a person waiting at the station the arrival of a standalone vehicle, the exact location where the vehicle will park. The position of the stopping mark must be such that it respects the minimum lateral clearance of the road and thus does not disturb or pose a risk to other vehicles (as defined by the traffic regulations). When the autonomous vehicle is parked over the stopping mark (302), its starting location sensor (203) must have a good line of sight of the starting location mark (303) in order to accurately measure the starting location.

[040] The starting location mark (303) may be, for example, a pole, a sign, a rectangle painted on a wall or even the floor so that it can be easily detected by the mounted starting location sensor (203). in the vehicle. Figures 4C and 4D show two examples of mounting initial location markings 421 and 422 printed on a signpost 420 formed of rectangles arranged to create high contrast regions, whereas the Vehicle's initial location sensor (203) is a camera. In another example, the vehicle may have a LASER sweep sensor as the start location sensor (203) and use a set of raised markers or even trees as the start location mark (105). In examples of multiple stopping points (Figures 3B and 3C), each starting location mark should be different from the other to allow the vehicle to identify which stop it is at, such as marks 421 and 422. . In another example of mounting, initial location marks painted in different colors may be used to allow differentiation by the vehicle. 3. Examples of Virtual Track Guiding Methods [041] Like a real track, the virtual track has the function of keeping the vehicle centered with the carriageway. The difference is that the virtual track does not exert mechanical force directly on the vehicle to change its direction to bring it back to the center. Instead, the virtual track serves only as a reference that allows the vehicle to measure how far it is from the center of the haul road. For example, a vehicle-mounted sensor can be used to measure this deviation and the central computer system automatically controls the steering wheel to bring the vehicle back to the center of the virtual track.

[042] The mounting example in Figure 5 shows a virtual rail mounting technique that does not require the installation of any special markers along the highway. Instead, this technique takes advantage of the available road sign information itself, such as horizontal lanes (503) or signs (501), to locate the vehicle on a virtual track map, with some degree of uncertainty. Any fixed object that is detectable by the track tracking sensor 508, such as a roadside tree 502, can be considered a locating object and taken advantage of in the vehicle locating process. The more locating objects along the road, the less uncertainty of the vehicle's location, and consequently the greater the reliability of autonomous driving. In order for the vehicle to be able to drive safely over the entire mapped area, location uncertainty should never be above limit values, such as 10cm lateral error and 2m longitudinal error, depending on operating speed and of the radius of curvature of the curves.

[043] The virtual track map is stored in computer memory (506), and can describe important road features such as its side boundaries (505), allowable speed for each stretch, track position (504), station position and the position of the locating objects. The virtual rail map can be produced by any standalone vehicle being driven by a human driver along any road route that starts at one station and ends at another. During this mapping trip, the information generated by the dead reckoning unit 509, such as encoders, gyroscopes and accelerometers can be merged with global location sensor information 511 to construct the path the vehicle has traveled. Simultaneously, the location object sensor (508) detects the location objects (501), (502), (503) present on the road and records them on the map according to the vehicle's current position. The speed at which the vehicle was driven at each point on the road can also be recorded and then served as a reference speed for autonomous operation on a transport mission.

[044] For example, a mapping operation begins with the precise initial location procedure of the vehicle, done within a station, when the initial location sensor (510), which depending on the mount may be the same object sensor as location (508), detects the initial location mark (105). This procedure can be, for example, the same as the “rail connection” procedure described in (pg19-1), and ensures that the new virtual rail leg starts at a known point of a station (stopping point) and thus is precisely concatenated to the virtual track that arrives at that same point. Following this methodology, a map of a complete virtual railway, with rails and stations, can be constructed by concatenating one-to-one mapped virtual rails.

[045] The path the vehicle travels during mapping can be considered directly as the virtual trail within the map, as it represents a free, road-centered path that if autonomously reproduced correctly will be able to drive the vehicle. from one station to another safely. If during the mapping trip a sudden maneuver has occurred, such as to deflect an unexpected obstacle, the virtual track can later be corrected in a specific program, where adjustments can be made for yaw removal, path smoothing, addition lateral displacements, modification of reference speeds for each section, and indication of parallel lateral lanes and shoulder.

[046] Autonomous driving of the vehicle along a path between two stations is by following a virtual track described in a virtual track map. Usually this process requires the vehicle to be able to determine with high reliability its location on the map and thereby calculate, for example, the lateral deviation from the track. In an assembly example, a probabilistic particle filter-based location technique may use information received from the vehicle's track tracking sensor to calculate the vehicle's probability distribution function (eg, x, y, thet). on the map at any moment. Still in this example, the location program may use the probability distribution function to calculate the reliability of the location and, for example, decide whether the vehicle can continue to operate in standalone mode or not. For example, if due to rain, visibility conditions worsen and consequently the reading of the track tracking sensor is impaired, it may occur that the location reliability level falls below a safety limit and therefore the vehicle is not be better able to drive yourself. In this case, the steering can be taken by a spare human driver, if any, or the vehicle itself performs an automatic safety stop maneuver, such as slowing down, stopping on the side, turning on the warning light and sending a text message to the destination requesting manual redemption.

[047] Once the vehicle knows its position on the map with high reliability and the map contains the position information of each point of the virtual track (which must also have been mapped with high reliability) then the vehicle is perfectly able to operate in standalone mode. In calculating the vehicle's lateral deviation from the rail, the central computing system may use this information to calculate the steering wheel positioning angle required to correct this deviation, and to send a command to the steering wheel drive unit (507).

Figures 6A and 6B show an example of mounting a virtual rail system based on magnetic markers installed under the highway pavement. The magnetic field sensor (602) captures the magnetic field generated by the installed magnets (603) and transmits to the central computing system (601) to calculate the deviation from the rail line (605). When there is a deviation, the steering computer can calculate, for example, the steering wheel turning angle required to bring the vehicle back to the center of the tracks. This information may be transmitted to a flywheel drive unit 604, for example a servo motor, which applies the force necessary to rotate the flywheel to the computer controlled position. For magnetic field detection, for example, a set of hall effect sensors or coils can be used. Each individual sensor can be mounted side by side so that the sensor that passes exactly over a magnet generates the most intense signal, thus indicating the lateral deviation of the vehicle. Figures 5C and 5D show another example of virtual track assembly that uses a painted floor strip as a guide element. An optical sensor (612), such as a camera, captures the road image and a computer (601) running a computer vision program calculates the vehicle's lateral deviation from the lane. Additionally, the same computer can further calculate the steering wheel turning angle required to bring the vehicle back to the center of the tracks and send this information to the steering wheel drive unit, for example a servo motor, which applies the force. required to rotate the steering wheel to the commanded position.

[049] Both magnetic guiding and painted lane guidance have the disadvantage of requiring the virtual rail to be constructed by installing signaling devices (magnets or special painted lanes) along the highways, which is expensive and expensive. brings the inconvenience of bans during installation. Another disadvantage is the permanent need to maintain this signaling, which increases the cost of operating this system. The great advantage of these two systems is the great reliability and relative simplicity of the programs that run on the vehicle's computer. 4. Example of obstacle detection sensors.

[050] Figure 7 shows an example of an obstacle sensor array consisting of three cameras and two simple LASER scan sensors. The two sensors (702), which may be two rear-facing cameras, have the function of covering the vehicle sides (711) to detect other vehicles approaching from behind. This type of detection is used in situations, for example, where the autonomous vehicle wants to change lanes or leave a station to enter the highway. The two LASER scan sensors 703, for example, LIDAR, are responsible for covering the front side of the sides 712, useful for detecting, for example, people or animals approaching the front of the vehicle from the sides and with This will prevent possible road kill. Depending on the need, a larger number of side LASER scan sensors 703 can be used to eliminate any blind spots around the vehicle, or a sensor with a coverage angle large enough to cover the entire side. The sensor 701 is responsible for covering the front region of the vehicle 710 and detecting, for example, the distance of the vehicle ahead, or if there is an obstacle or impediment in the middle of the track. In another mounting example, the front region cover (710) can also be made with the help of a LASER scan sensor and a microwave radar.

[051] The information generated by this set of obstacle sensors can be processed separately by a computer dedicated to this task so that it does not overload the vehicle's main computer responsible for controlling the autonomous transport mission. In another assembly example, a single computer with high processing capacity can manage mission control tasks, obstacle detection, and track tracking.

[052] When an obstacle is detected by the obstacle sensing system, the obstacle detection program module classifies the obstacle by type (eg car, pedestrian), calculates its location and speed. This information is transmitted to the autonomous mission control module that is driving the vehicle so that the correct reaction can be taken. For example, if an approaching pedestrian in sector 712 is detected moving at a certain speed towards the front of the vehicle, mission control can use this information in conjunction with the current vehicle speed information to calculate the risk of possible running over. If the risk is low, the vehicle simply continues on the virtual track, but if the risk is high, the vehicle may react by performing a detour or stop maneuver. Another example of reaction to an obstacle is the case of the front sector sensor (710) detecting a vehicle driving in front of it at a speed slower than programmed in the transport mission. An example of a reaction may be speed reduction to keep the autonomous vehicle behind the other vehicle at a constant safety distance. Another possibility of reaction may be the overtaking maneuver, where the permission to change lanes is made by detecting no vehicles approaching from the rear side in the sector field of view (711), for example. 5. Example of autonomous transport operation and mission [053] An example of autonomous transport operation is presented in figure 10. In this example, the operation starts with the driver loading the mission file (801), which defines, for example, For example, which way the vehicle will take, at which stations it will stop and at what speed to cruise for each leg of the trip. Then the vehicle awaits the command to start the execution of the mission (802), which can be given through the user interface (206). Upon receiving this command, and if the vehicle is already precisely located on the map (“rail-connected”), the vehicle then begins carrying out the transport mission (803). If the vehicle is not yet connected, it performs the connection procedure (804) and then proceeds with the mission. In the event that the vehicle fails during the connection procedure, the mission cannot be performed and the vehicle can signal the driver to this fault condition (805) and / or can send a message to a monitoring center.

[054] Figure 11 shows an example of a procedure that can be used to connect the vehicle to the virtual railway. The procedure begins with the driver driving the vehicle to a station and parking within the region indicated by the stop mark (104). The driver then gives the command "rail connection", which can be done, for example, through the user interface (206). In response to this command, the vehicle computer uses the global location information provided by the global location sensor to calculate which station it is at (903). Following this process, the vehicle uses the home location sensor to detect the home location mark (105) and accurately calculate its home position (904). If no failure occurs in steps (903) and (904) the vehicle will be able to perform an autonomous transport mission. If a fault occurs, this condition is signaled (906) to the driver by means of an alarm, for example.

[055] Figure 10A shows an example of a simple autonomous transport mission where the vehicle must carry a cargo from one station to another. In this example, the vehicle starts the mission from station # 6, following the virtual tracks until it enters extension # 4 (1001). Before starting to move, the vehicle may use the front obstacle sensor (710) to check that there is no obstruction ahead, and before entering the highway (access extension # 4), use the sector obstacle sensor (711) to check that there is no other vehicle approaching. Upon entering the highway, the vehicle follows the virtual tracks with the cruise speed specified in the mission until it approaches station # 2 (1002). During autonomous travel, the vehicle uses obstacle sensors to avoid collisions and accidents as described in (p. 16-22). Finally, as you approach the destination station, the vehicle signals with the arrow lights that it will leave the lane, slows down to the appropriate maneuver value, enters the station and stops smoothly over the stop mark (1003). After stopping, the vehicle signals that the mission has been completed and the vehicle is in manual mode, ready to be unloaded (1004). Figure 10B shows another example of a mission where the vehicle departs from station # 6 (1011), stops loading and unloading at station # 5 (1013), waits for the command to proceed (1014), returns to extension (1016). and drives to station # 4 (1017) where it parks (1018) and signals the end of mission (1019). In another example, a mission may be programmed so that the vehicle enters and stops at all stations along the route. In a preferred mode of operation, the command for the vehicle to proceed with the journey after stopping at an intermediate station may be given by the driver of the vehicle himself, who acts as a kind of driver overseeing the journey, but may also be given, for example. by a station worker loading / unloading the vehicle.

Claims (10)

1. AUTONOMOUS ROAD TRANSPORT SYSTEM characterized by comprising at least one autonomous vehicle equipped with electronic control system that allows to make autonomous trips between stations guided by virtual rails; with at least two stations serving as a stopping point for autonomous vehicles to load or unload goods or persons; and with a virtual track system that contains natural or unnatural locating objects that can be used to guide the autonomous vehicle along routes between stations. 2. AUTONOMOUS ROAD TRANSPORT SYSTEM according to claim 1, characterized in that it comprises a track tracking system responsible for keeping the vehicle centered on the road in the routes between the stations; it contains an obstacle detection system, which is responsible for detecting obstacles that are in the vicinity of the vehicle and which may present a risk of collision; it contains a mission execution controller responsible for coordinating the progress of a mission, managing operator commands, vehicle speed and performing collision avoidance maneuvers, and it contains an operator interface through which the operator oversees autonomous vehicle operation . 3. AUTONOMOUS ROAD TRANSPORT SYSTEM according to claim 1, characterized in that it comprises a station having at least one stop mark which has the function of marking the exact place where the vehicle is to stop within a station and containing at least one station. least one start location mark, which can be detected by the autonomous vehicle start location sensor when it is standing over the stop mark. 4. AUTONOMOUS ROAD TRANSPORT SYSTEM according to claim 2, characterized in that it comprises a track tracking system having a virtual railway map in digital format with information describing the position of the tracks, the position of the locating objects, the position of the stations with their respective stop marks the existence of a shoulder lane; that contains a global location sensor that allows you to identify which map station the vehicle is on; containing a home location sensor configured to detect the home location mark of a station; containing a track tracking sensor configured to detect highway location objects used by the track tracking module to accurately determine the vehicle's position relative to the virtual track; which contains a dead reckoning unit to calculate incremental vehicle offsets; which contains an initial location program module responsible for determining the vehicle's initial position on the map with high precision in the procedure referred to in the Rail Link Descriptive Report (pg19-1); containing a locating program module configured to precisely determine the vehicle's position relative to the rail from the locating object detection information, incremental offsets and rail map, and containing a steering control program module configured to calculate the lateral deviation of the vehicle from the tracks and to calculate the steering angle of the steering wheel needed to correct this deviation. 5. AUTONOMOUS ROAD TRANSPORT SYSTEM according to claim 2, comprising an obstacle detection system having an obstacle sensor configured to detect obstacles close to the vehicle and containing an obstacle detection program module configured in to process information from obstacle sensors and identify obstacles that pose a risk of collision. 6. AUTONOMOUS ROAD TRANSPORT SYSTEM according to claim 2, characterized by a mission execution controller having a mission control program module that uses the information generated by the track tracking system and obstacle detection system to Deciding how to act on the car's controls (steering wheel, brake, throttle and lights) so that the vehicle performs the mission correctly while avoiding collisions. 7. AUTONOMOUS ROAD TRANSPORT SYSTEM according to claim 2, characterized by an operator interface through which the operator supervises the operation of the autonomous vehicle and can control and monitor the operation of the autonomous vehicle. Autonomous road transport system according to claim 4, characterized by an initial location sensor wherein the sensor is a camcorder or a LASER scan sensor. Autonomous road transport system according to claim 4, characterized by a track tracking sensor wherein the sensor is a camcorder or a LASER scan sensor. Autonomous road transport system according to claim 5, characterized by an obstacle sensor in which the sensor consists of one or more video cameras or one or more LASER sensors.