Invented by Karl Iagnemma, Motional AD LLC

The market for route planning for an autonomous vehicle is rapidly expanding as the demand for self-driving cars continues to grow. With advancements in technology and the potential for increased safety and efficiency, autonomous vehicles are becoming an attractive option for consumers and businesses alike. However, for these vehicles to operate successfully, they require sophisticated route planning systems that can navigate complex road networks and make real-time decisions. One of the key factors driving the market for route planning in autonomous vehicles is the need for efficient and optimized routes. These vehicles rely on accurate and up-to-date mapping data to determine the most efficient path to their destination. Route planning algorithms analyze various factors such as traffic conditions, road closures, and weather conditions to calculate the optimal route. By minimizing travel time and fuel consumption, these systems can help reduce costs and improve overall efficiency. Another important aspect of route planning for autonomous vehicles is safety. These vehicles need to be able to navigate safely through various road conditions and handle unexpected situations. Route planning systems take into account factors such as speed limits, traffic rules, and road hazards to ensure a safe journey. Additionally, these systems can also incorporate real-time data from other vehicles and infrastructure to make informed decisions and avoid potential accidents. The market for route planning in autonomous vehicles is not limited to personal transportation. Industries such as logistics and delivery services can greatly benefit from these systems. With the ability to optimize routes and make real-time adjustments, autonomous vehicles can streamline delivery operations and reduce delivery times. This can lead to increased customer satisfaction and cost savings for businesses. Furthermore, the market for route planning in autonomous vehicles is also driven by the need for connectivity and integration with other smart devices and systems. These vehicles can communicate with traffic management systems, infrastructure sensors, and other vehicles to exchange information and improve overall traffic flow. This integration allows for more efficient route planning and can help reduce congestion on the roads. As the market for autonomous vehicles continues to grow, so does the demand for advanced route planning systems. Companies specializing in mapping and navigation technologies are investing heavily in research and development to create more sophisticated algorithms and mapping data. Additionally, collaborations between automakers, technology companies, and mapping providers are becoming more common to ensure seamless integration and compatibility. In conclusion, the market for route planning in autonomous vehicles is expanding rapidly as the demand for self-driving cars increases. These vehicles require sophisticated systems that can optimize routes, ensure safety, and integrate with other smart devices and systems. With advancements in technology and increased investments in research and development, the future of route planning for autonomous vehicles looks promising.

The Motional AD LLC invention works as follows

Among other things… a determination is done of the ability of an autonomous car to travel safely or robustly a road feature, a segment of road or a route being considered by the autonomous vehicle at a certain time or range of dates. The route root is mapped to the properties of road network data. The computer will eliminate the road segment, road feature, or route if it determines that the vehicle cannot safely or robustly travel the segment, road feature, or route. The computer’s determination is based on the properties of the environment where the autonomous vehicle travels.

Background for Route Planning for an Autonomous Vehicle

This description is about route planning for an automated vehicle.

An autonomous car can drive without human interference during a part of the journey or for the entire journey.

An autonomous vehicle is equipped with sensors, actuators and computers to automatically generate and follow routes in the environment. Some autonomous vehicles are equipped with wireless two-way communications to allow them to communicate with command centers located remotely, which may have human monitors on site, or to access information and data stored in cloud services, as well as to communicate to emergency services.

As shown in FIG. In a typical autonomous vehicle 10 use, the desired goal position 12 can be identified by a number of different methods. The goal position may be specified by a rider (who may be, for example, an owner of the vehicle or a passenger in a mobility-as-a-service ?robo-taxi? application). The goal position can be determined by an algorithm. (For example, the algorithm may run on a cloud-based server and be tasked with optimizing locations for a fleet autonomous vehicles in order to minimize rider wait time when hailing a robot-taxi). The goal position can be determined by a specific process.

Given the desired goal position, the routing algorithm 20 will determine a route through the environment to get from the current position of the vehicle (16) to the desired position (12). This process is sometimes called ‘route planning.’ “In some implementations, the route is made up of a series connected segments of streets, roads and highways. (We sometimes call these segments road segments or just segments.)

Routing algorithm typically operates by analyzing information about the road network. Road network information is typically a digital representation that includes information such as the type, structure, connectivity and other relevant data about the network. A road network can be represented by a series connected road segments. In addition to identifying the connectivity between road segments the information may also contain information about physical and conceptual characteristics of each segment. This includes but is not limited to: geographic location, road number or name, road width and length, speed limit and direction of travel.

The routing algorithm typically identifies a candidate route 22 that connects the current position with the desired position. The best or optimal route 14 is identified by using algorithms (such Dijkstra?s algorithm, A*, D* and others), which identify the route that minimizes the specified cost. The cost of a route is usually a function one or more factors, such as the speed limit, the traffic conditions and the distance traveled. The algorithm can identify several good routes that are presented to the rider or other person (for example, an operator in a remote location), for approval or selection. The optimal route can be given to a vehicle trajectory control module 28 that has the purpose of guiding the vehicle to the desired goal along the optimal path.

As shown in FIG. The road network information is typically stored in a database 30, which is kept on a centrally-accessible server 32. This database can be updated frequently (e.g. 1 Hz). The network information is available either on demand (e.g. requested by the vehicle) or by server.

Road Network Information can be associated with temporal information, to allow descriptions of traffic rules or parking rules or other effects which are time-dependent (e.g. a segment of road that does not permit parking during normal business hours or weekends, for instance), or to include expected travel times along a particular road segment at certain times of the day (e.g. during rush hour).

In general, an aspect determines the ability of an autonomous car to travel safely or robustly a road feature, a road segment, or a route, that is being considered by the autonomous vehicle, as a result of a certain time or range of dates. The route root is based on the properties of information stored about road networks. The computer will eliminate the road feature, road segment, or route if it determines that the autonomous vehicle cannot safely or robustly travel the route. The computer’s determination is based on the properties of the environment where the autonomous vehicle travels.

Implementations can include a single feature or a combination. The environment also includes road features. Navigability of the autonomous vehicle is one of the properties of the environment. The spatial characteristics of road features are included in the properties of the environment. Connectivity characteristics of road features are included in the properties of environment. The spatial orientation of road features is one of the properties of an environment. Locations of roadwork or accidents are included in the properties of an environment. The roughness of the road surface is one of the properties of the environment. Curvature slope is one of the properties of environment that can affect visibility. The characteristics of road markings are included in the properties of environment. Physical navigation challenges associated with road features are part of the properties of the environmental. The computer determines the ability of an autonomous vehicle to travel safely or robustly each of a series of road features, road segments or routes.

The characteristics of the sensors in the vehicle will determine whether the autonomous vehicle can safely or robustly navigate a road feature, a road segment, or a route. These characteristics can include a level of performance that is actual or estimated based on current or future conditions. The computer calculates the capability of the autonomous vehicle at a certain time. A route planning process determines two or more possible routes. Software processes determine the ability of an autonomous vehicle to travel safely and robustly a road feature, a segment of road or a route. Software processes can include the processing of data collected by sensors mounted on the vehicle. Software processes include planning of motion. Software processes include decision making. Software processes include vehicle movement control. “The characteristics include a level of performance that is actual or estimated based on current or future conditions.

The following can be used to express “These and other features, implementations and advantages” in other ways: as methods, systems components, apparatuses, program products, business methods, means and steps of performing functions and more.

The following description will reveal other aspects, features and implementations.

DESCRIPTION

FIGS. “FIGS.

FIGS. “FIGS.

FIG. “FIG.

For route planning for human-piloted cars, it is assumed that a route from a present position to a destination position composed of road segments that are connected is a safe route. This assumption, however, may not hold true for autonomous vehicles. Due to their capabilities and the characteristics of certain road features, autonomous vehicles may be unable to safely navigate some road segments, intersections or other geographical regions. Autonomous vehicles may also not be able navigate certain road features safely at certain times, seasons, or weather conditions.

In FIGS. 3 and 10, we show an example of where sensors and software are located in a car and on a cloud-based database and server. “Figures 3 and 10” show an example of the physical locations of sensors and software processes in a vehicle, as well as at a cloud-based server.

Sensors & Software Processes

Click here to view the patent on Google Patents.