Method and arrangement for monitoring and adapting the performance of a fusion system of an autonomous vehicle

Patent metadata

Field	Value
Publication number	US-2017297571-A1
Application number	US-201715467297-A
Country	US
Kind code	A1
Filing date	Mar 23, 2017
Priority date	Apr 14, 2016
Publication date	Oct 19, 2017
Grant date	—

Abstract

Disclosed herein is a method and arrangement for monitoring and adapting the performance of a fusion system of an autonomous road vehicle. A drivable area is determined by combining a localization function and high density map data. Information on surrounding objects is determined, comprising determining the localization and classifying the surrounding objects, estimating their physical property states, and assigning them extension descriptions. Information on the drivable area and surrounding objects is condensed into observed areas, monitored by sensors of the environmental perception function with a predetermined degree of certainty, and prioritized objects, represented by classes, state estimates and extension descriptions. Having the fusion system monitor itself retrospectively by evaluating its current determinations of drivable area, prioritized objects and observed areas against its previous determinations thereof, and if a previous determination differ more than a predetermined amount from a current determination, adapting the fusion system to account for that discrepancy.

First claim

What is claimed is:

1 . A method for monitoring and adapting performance of a fusion system of an autonomous road vehicle equipped with an autonomous drive system comprising a positioning system and multiple sensors for supervising an internal state of the autonomous road vehicle and a surrounding environment, the method comprising:
providing a localization function of the fusion system by fusing data from the positioning system and the sensors; providing an environmental perception function of the fusion system by fusing data from the sensors supervising the surrounding environment; implementing the localization and environmental perception functions of the fusion system using parallel algorithms based on alternative concepts and using different sets of sensor information; determining a drivable area in front of the autonomous road vehicle by combining the localization function and high density map data for a current position; determining information on surrounding objects, the information determination comprising determining localization of the surrounding objects, classifying the surrounding objects, estimating physical property states of the surrounding objects, and assigning the surrounding objects extension descriptions and a highly automated drive confidence; condensing information on the drivable area and surrounding objects into observed areas and prioritized objects, specifying observed areas as a volume around the autonomous road vehicle that has been monitored by the sensors with a predetermined degree of certainty, and representing prioritized objects by classes, state estimates, extension descriptions and highly automated drive confidences; and having the fusion system monitor itself retrospectively by evaluating its current determinations of drivable area, prioritized objects and observed areas against its previous determinations thereof, and if a previous determination differs more than a predetermined amount from a current determination, adapting the fusion system to account for that discrepancy.

2 . The method according to claim 1 wherein the drivable area in front of the autonomous road vehicle is determined as an area that belongs to a road travelled, which, given sensor and algorithmic uncertainties, is considered safe with a predetermined probability.

3 . The method according to claim 1 wherein the determination of information on surrounding objects comprises classifying surrounding objects with respect to object types, estimating physical property states of surrounding objects including their positions, velocities and accelerations, and assigning the surrounding objects extension descriptions that include sensor uncertainty descriptions.

4 . The method according to claim 3 further comprising determining if a current determination of a class of a prioritized object differs from a previously determined class of that prioritized object, and if so adapting the fusion system by increasing a sensor uncertainty description associated with that previously determined class.

5 . The method according to claim 3 further comprising determining if a current localization determination of a prioritized object differ from a previous localization determination of that prioritized object more than what a reasonable estimate based on its previous representation would indicate but indicates a localization in a predetermined close vicinity thereto, and if so adapting the fusion system by increasing a sensor uncertainty description associated with that prioritized object.

6 . The method according to claim 1 further comprising determining if a current localization determination of a prioritized object differ from a previous localization determination of that prioritized object more than what a reasonable estimate based on its previous representation would indicate and not indicates a localization in a predetermined close vicinity thereto, and if so adapting the fusion system by distrusting the fusion system and initiating hand-over of control of the autonomous road vehicle from the autonomous drive system to a driver thereof.

7 . The method according to claim 1 further comprising determining if a current determination of drivable area indicates a different extension of the drivable area than a previous determination thereof, and if so adapting the fusion system such that future determinations of drivable area are made based on the current determination.

8 . The method according to claim 7 further comprising determining if a current determination of drivable area indicates a more narrow width of the drivable area than a previous determination thereof, and if so adapting the fusion system such that future determinations of drivable area are made based on a more narrow width of the drivable area than that of the previous determination thereof.

9 . The method according to claim 7 further comprising determining if a current determination of drivable area indicates a less narrow width of the drivable area than a previous determination thereof, and if so adapting the fusion system such that future determinations of drivable area are made based on a less narrow width of the drivable area than that of the previous determination thereof.

10 . The method according to claim 1 further comprising determining if a current determination of drivable area indicates that the previously determined drivable area was not drivable at all, and if so adapting the fusion system by distrusting the localization function and initiating hand-over of control of the autonomous road vehicle from the autonomous drive system to a driver thereof.

11 . The method according to claim 1 further comprising determining if a current determination of surrounding objects indicates that a new object has appeared in an area where it should have been determined previously, and if so adapting the fusion system by distrusting the fusion system and initiating hand-over of control of the autonomous road vehicle from the autonomous drive system to a driver thereof.

12 . A fusion system for use in an autonomous road vehicle equipped with an autonomous drive system comprising a positioning system and multiple sensors for supervising an internal state of the autonomous road vehicle and a surrounding environment, the fusion system comprising:
a localization function provided by fusing data from the positioning system and the sensors; an environmental perception function provided by fusing data from the sensors supervising the surrounding environment; wherein the localization and environmental perception functions of the fusion system are implemented using parallel algorithms based on alternative concepts and using different sets of sensor information; wherein the fusion system configured to
determine a drivable area in front of the autonomous road vehicle by combining the localization function and high density map data for a current position;
determine information on surrounding objects, the information determination comprising determining localization of the surrounding objects, classifying the surrounding objects, estimating physical property states of the surrounding objects, and assigning the surrounding objects extension descriptions and a highly automated drive confidence;
condense information on the drivable area and surrounding objects into observed areas and prioritized objects, wherein observed areas are specified as a volume around the autonomous road vehicle that has been monitored by the sensors with a predetermined degree of certainty, and prioritized objects are represented by classes, state estimates, extension descriptions and highly automated drive confidences; and
monitor itself retros

Assignees

• Volvo Car Corp

Inventors

• Sorstedt Joakim Lin
• Svensson Daniel

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G01S13/86

  Combinations of radar systems with non-radar systems, e.g. sonar, direction finder · CPC title

• B60W2554/80

  Spatial relation or speed relative to objects · CPC title

• B60W2754/10

  Spatial relation or speed relative to objects · CPC title

• B60W2556/50

  of positioning data, e.g. GPS [Global Positioning System] data · CPC title