Invented by David Ian Franklin Ferguson, Waymo LLC

The market for determining changes in driving environments based on vehicle behavior is rapidly growing, thanks to advancements in technology and the increasing demand for safer and more efficient transportation systems. This market encompasses various industries, including automotive, transportation, and data analytics, and is driven by the need for real-time monitoring and analysis of driving conditions. One of the key drivers of this market is the rise of connected vehicles. Modern cars are equipped with a wide range of sensors and communication technologies that enable them to collect and transmit data about their surroundings. This data includes information about road conditions, weather conditions, traffic patterns, and even the behavior of other vehicles on the road. By analyzing this data, companies can gain valuable insights into the driving environment and make informed decisions to improve safety and efficiency. Another factor contributing to the growth of this market is the increasing focus on autonomous driving. As self-driving technology continues to advance, it becomes crucial to accurately determine changes in the driving environment to ensure the safe operation of autonomous vehicles. By analyzing vehicle behavior and the surrounding environment, autonomous vehicles can make real-time adjustments to their driving strategies, such as changing lanes, adjusting speed, or avoiding obstacles. The market for determining changes in driving environments based on vehicle behavior also presents significant opportunities for data analytics companies. The sheer volume of data generated by connected vehicles provides a wealth of information that can be analyzed to identify patterns, trends, and anomalies. This data can be used to develop predictive models, optimize traffic flow, and even detect potential hazards before they occur. Furthermore, this market has the potential to revolutionize transportation systems and improve road safety. By accurately determining changes in driving environments, companies can develop advanced driver assistance systems (ADAS) that provide real-time warnings and alerts to drivers. These systems can help prevent accidents, reduce congestion, and enhance overall driving experience. However, there are also challenges that need to be addressed in this market. One of the main challenges is ensuring the privacy and security of the data collected from connected vehicles. As vehicles become more connected, there is a risk of unauthorized access and misuse of sensitive information. Companies operating in this market need to invest in robust cybersecurity measures to protect the data and maintain consumer trust. In conclusion, the market for determining changes in driving environments based on vehicle behavior is experiencing significant growth and presents numerous opportunities for various industries. The rise of connected vehicles and the increasing focus on autonomous driving are driving the demand for real-time monitoring and analysis of driving conditions. This market has the potential to revolutionize transportation systems, improve road safety, and enhance overall driving experience. However, addressing privacy and security concerns is crucial to ensure the success and sustainability of this market.

The Waymo LLC invention works as follows

A method and an apparatus are provided to determine whether a driving situation has changed in relation to previously stored data about the driving situation. The apparatus can include an autonomous driving system that detects one or more vehicles within the driving environment and calculates corresponding trajectory for these detected vehicles. The autonomous computer system can then compare the determined trajectory to a predicted trajectory of a hypothetical car in the driving area. The autonomous driving computer system can determine if the driving conditions have changed or a probability of the changes based on the comparison.

Background for Determining the changes in driving environments based on vehicle behaviour

An autonomous car may use different computing systems to help transport passengers from one place to another. The autonomous vehicle can also require an operator’s input, whether it is a driver, pilot or passenger. Autopilot systems and other autonomous systems may only be used when the system is engaged. This allows the operator to switch between a manual mode, where the driver has a great deal of control, to an autonomous mode, where the vehicle drives itself.

The autonomous vehicle can be equipped with different types of sensors to detect objects within its environment. The autonomous vehicles can include sensors such as sonars, radars, cameras and other sensors to scan and record information from their environment. The sensor data from one or several of these sensors can be used to detect and identify objects, their characteristics (position, size, direction, speed, etc.). This detection and identification function is critical for the safe operation the autonomous vehicle.

To navigate in an environment with confidence and precision, the autonomous vehicle can rely on the previously stored electronic representation (e.g. a road, a highway etc.). The electronic representation of an environment can be considered as a “map”. This includes features such as lane markings and edges, concrete k-rail barriers, lane dividers on roads, traffic cones for safety, road medians etc. “The autonomous vehicle can store maps for both simple and complex environments.

There are instances when these previously stored maps can be inaccurate or out of date. There may be construction going on in an area or a road accident. The lanes on the road may shift relative to the position indicated in the previous stored map. The autonomous vehicle should be able identify these changes on the road in such situations.

A method and apparatus are disclosed. In one embodiment, an apparatus is provided with a sensor that detects a first car in a driving situation, as well as a computer-readable storage device for detailed map data for the driving scenario. The detailed map data includes information about the road the vehicle travels on, and the first state of the vehicle. The apparatus can also include a computer-readable memory in communication with both the sensor and the processor. The processor can be configured to receive sensor data from the sensor based upon the detection of the first vehicle within the driving area, determine first state information using the received sensor data, determine a trajectory based upon the first state, determine an expected path based on detailed map information and determine whether the driving environment is different by comparing this determined expected path with the determined first route.

In another embodiment, the processor is configured to calculate a deviation value by comparing a determined expected trajectory with a determined first trajectory. The processor then determines that driving conditions have changed if the deviation value exceeds the deviation threshold.

In a further embodiment, the deviation metric is a maximum deviation value that represents the maximum difference between the first trajectory determined and the expected trajectory determined.

In yet another embodiment, the determined deviation value is an average signed value of deviation, where the average signed value represents a magnitude and a direction of a deviation between the first trajectory determined and the expected trajectory determined.

In a still further embodiment, the first trajectory determined by the apparatus is an average trajectory. The average trajectory has been averaged across a predetermined period of time.

In another embodiment, the device determines the expected trajectory based on the centerline of the roadway corresponding to detailed map information.

In a further embodiment, the computer-readable storage stores a probabilistic model that calculates a probability of the driving situation changing in relation to the detailed maps information, based upon at least one deviation value determined by comparing the first trajectory determined with the expected trajectory determined, and a probabilistic function that computes this probability based on a model of probability. The processor can also be configured to determine, based on the probabilities function, whether the driving environment is different from the detailed map data.

In yet another embodiment, the probability is selected from a number of different probability models. The processor can be configured to choose the probability model based on the first geographical location.

In a still further embodiment, the first trajectory determined by the apparatus comprises a plurality trajectories. Each of these trajectories corresponds to a particular vehicle in the driving area. The processor can be configured to further consolidate the plurality trajectories into the first trajectory determined by the apparatus based on a minimum of one consolidation factor.

The processor can be configured in another embodiment to calculate a consolidated quality value of the first trajectory. This value represents the quality of that trajectory. Based on this value, it is possible to determine whether the driving environment changed compared to the detailed map data.

In one embodiment, the method can include detecting a first car in a driving area with a sensor on an autonomous vehicle and receiving sensor information from the sensor based on the fact that the first car was detected in the driving area. The method can also include determining with the processor the first information from the sensor information. This first information may identify at least one of the following: position, speed or direction of travel of the first vehicle. The method can also include determining a first path based upon the first state information and determining an expected path based upon detailed map data, where the detailed map data includes information about the driving conditions in which the vehicle is traveling. The method can also include determining with the processor that the driving conditions have changed by comparing a determined expected trajectory to a determined first trajectory.

In another embodiment, the method can include determining with the processor a deviation value by comparing a determined expected trajectory with a determined first trajectory and determining with the processor that the driving conditions have changed when the deviation value exceeds the deviation threshold.

In a further embodiment, the calculated deviation metric includes a maximum deviation value that represents a maximum difference in the first trajectory determined and the expected trajectory determined.

In yet another embodiment, the calculated deviation metric values comprises an averaged signed deviation value. The averaged signed deviation value represents a magnitude and a direction of a different between the first trajectory determined and the expected trajectory determined.

In a still further embodiment, the first trajectory determined is an average trajectory. The average trajectory has been averaged across a predetermined period of time.

In another embodiment, the expected trajectory is determined based on the centerline of a road that corresponds to the detailed map data.

In a further embodiment, the method comprises determining with the processor a probabilistic function that the driving information has changed in relation to the detailed maps information, wherein the function defines the probability of a driving information change relative to detailed maps information using a model of probability, and the model of probability defines a likelihood that the driving information has changed with respect to detailed maps information on the basis of at least one deviation value calculated from a comparison of the first trajectory determined with the expected trajectory determined.

Click here to view the patent on Google Patents.