Obstacle detection for a rail vehicle

Abstract

A method for obstacle detection for a rail vehicle includes determining a monitoring region in front of the vehicle based on position and course of a track which the vehicle travels and position and dimensions of a monitoring profile of the vehicle. During the determination, sensor data for determining position and orientation of the vehicle are captured and position and course of the track which the vehicle travels are determined based on alignment of the determined position and orientation of the vehicle with digital map data. Objects in the vehicle environment are captured by sensors of the vehicle. Which objects are potential collision obstacles is determined by checking whether and to what extent they overlap with the monitoring region. A reaction to the objects categorized as potential obstacles is determined based on geometric and kinetic attributes of the objects. An obstacle detection device and a vehicle are also provided.

1 - 13 . (canceled) 
     
     
         14 . A method for obstacle detection for a rail vehicle, the method comprising the following steps:
 determining a monitoring region in front of the rail vehicle as a function of:
 a position and a course of a track on which the rail vehicle is travelling, and 
 a position and dimensions of a monitoring profile of the rail vehicle; 
   capturing sensor data for determining at least one of a position or an orientation of the rail vehicle;   determining the position and the course of the track on which the rail vehicle is travelling based on at least one of an alignment of the determined position or an orientation of the rail vehicle with digital map data;   capturing objects in an environment of the rail vehicle by using sensors;   determining which of the objects being potential collision obstacles by checking whether and to what extent the obstacles overlap with the monitoring region; and   determining a reaction to the objects categorized as potential obstacles, as a function of at least one of geometric or kinetic attributes of the objects.   
     
     
         15 . The method according to  claim 14 , which further comprises carrying out at least one of a calibration of the position or the orientation of the rail vehicle by:
 determining positional data from landmarks in an environment of the rail vehicle; and   comparing the positional data relating to the landmarks with positional data in the digital map data.   
     
     
         16 . The method according to  claim 14 , which further comprises providing the attributes of the objects as at least one attribute type including:
 approximate dimensions and an area of a respective object; or   information indicating a volume overlapped by the object; or   an interval from the rail vehicle to the object; or   dynamic attributes of the objects.   
     
     
         17 . The method according to  claim 15 , which further comprises providing the sensor data as:
 satellite navigation data; or   IMU data; or   speed sensor data; or   odometry data; or   infrastructure-based positional data; or   feature data detected in the environment of the rail vehicle for auto-location-finding, the feature data configured to be aligned with the digital map; or   sensor data relating to the environment of the rail vehicle for a specification of object attributes.   
     
     
         18 . The method according to  claim 17 , which further comprises providing the feature data for auto-location-finding as:
 landmarks; or   the track.   
     
     
         19 . The method according to  claim 14 , which further comprises combining sensor data from a plurality of sources to determine at least one of the position or the orientation of the rail vehicle in the digital map. 
     
     
         20 . The method according to  claim 15 , which further comprises carrying out the method by:
 using a Kalman filter or particle filter for filtering; or   using graph-based methods; or   using sensor-specific filtering; and   restricting a status of the rail vehicle to a movement of the rail vehicle along the track.   
     
     
         21 . The method according to  claim 14 , which further comprises carrying out the step of determining which objects are possible collision obstacles by:
 generating a point cloud of object points based on the sensor data relating to the environment from one or a plurality of sensors;   breaking the point cloud down into individual objects by clustering; and   determining an overlap region of individual objects with a sectional area of the monitoring region.   
     
     
         22 . The method according to  claim 21 , which further comprises carrying out the determination of the overlap region of the individual objects with the sectional area of the monitoring region by:
 determining a closest object point lying closest to the rail vehicle as measured along the track on which the rail vehicle is travelling;   grouping together all object points in a 2D plane by combining all projection planes running through each object point and being oriented orthogonally to the track, all of the object points of the point cloud being projected onto the projection plane of the closest object point; and   determining attributes of the individual objects based on a subset of the object points within the sectional area of the monitoring region.   
     
     
         23 . An obstacle detection device, comprising:
 a monitoring region determination unit for determining a monitoring region in front of a rail vehicle as a function of:
 at least one of a position or a course of a track on which the rail vehicle is travelling, and 
 at least one of a position or dimensions of a monitoring profile of the rail vehicle; 
   sensor data for determining at least one of a position or an orientation of the rail vehicle being captured;   at least one of the position or the course of the track on which the rail vehicle is travelling being determined based on at least one of an alignment of the determined position or the orientation of the rail vehicle with digital map data;   a sensor unit for sensory capture of objects in an environment of the rail vehicle;   an obstacle determination unit for determining which objects are potential collision obstacles by checking whether and to what extent the objects overlap with the monitoring region; and   a response determination unit for determining a reaction to the objects categorized as potential collision obstacles, as a function of at least one of geometric or kinetic attributes of the objects.   
     
     
         24 . A rail vehicle, comprising:
 an obstacle detection device according to claim  23 ; and   a control device for controlling a driving strategy of the rail vehicle as a function of the reaction determined by the obstacle detection device.   
     
     
         25 . A non-transitory computer program product, comprising a computer program configured to be loaded directly into a storage unit of a control device of a rail vehicle, with program sections for executing all steps of the method according to  claim 14  when the computer program is executed in the control device. 
     
     
         26 . A non-transitory computer-readable medium on which program sections configured to be executed by a computing unit are stored in order to execute all steps of the method according to  claim 14  upon the program sections being executed by the computing unit.