Method for determining an evasive path for a host vehicle

• US 10,410,521 B2
• Filed: 05/07/2015
• Issued: 09/10/2019
• Est. Priority Date: 05/08/2014
• Status: Active Grant

• Current Assignee: Volvo Car Corporation (AB Volvo), Volvo Truck Corporation (AB Volvo)
• Original Assignee: Volvo Car Corporation (AB Volvo)
• Inventors: Madas, David, Silvlin, Jonatan, Eidehall, Andreas, Sundstrom, Peter
• Primary Examiner(s): Shafi, Muhammad
• Application Number:
  US15/307,955
• Publication Number:
  US 20170053532A1
• Time in Patent Office: 1,587 Days
• Field of Search: 701117
• CPC Class Codes: B60W 10/18 including control of braking systems
• CPC Class Codes: B60W 10/20 including control of steering systems
• CPC Class Codes: B60W 2554/00 Input parameters relating to objects
• CPC Class Codes: B60W 2710/18 Braking system
• CPC Class Codes: B60W 2710/20 Steering systems
• CPC Class Codes: B60W 30/09 Taking automatic action to avoid collision, e.g. braking and steering
• CPC Class Codes: B60W 30/095 Predicting travel path or likelihood of collision
• CPC Class Codes: B60W 30/0956 the prediction being responsive to traffic or environmental parameters
• CPC Class Codes: B62D 15/0265 Automatic obstacle avoidance by steering
• CPC Class Codes: G01C 21/3415 Dynamic re-routing, e.g. recalculating the route when the user deviates from calculated route or after detecting real-time traffic data or accidents
• CPC Class Codes: G08G 1/163 involving continuous checking
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes

Abstract

The present invention relates to a method for determining an evasive path for a host vehicle (10), the method i.a. comprising: establishing (S1) a predicted traffic environment (12) of the host vehicle in a time-lateral position domain, the predicted traffic environment comprising an object prediction representation (16) of a traffic object; determining (S2) a start node (24) for the host vehicle; defining (S3) an end node (26) for the host vehicle; placing (S4) several boundary nodes (28) relative to the object prediction representation; setting (S5) node connections (30, 32, 34a-b, 36a-b, 38, 40, 42) between the start node, the boundary nodes, and the end node; and traversing (S6) the nodes using a graph-search algorithm in order to find a path from the start node to the end time with a lowest cost.

Claims

1. A method, comprising:

detecting a traffic object that is external to a host vehicle in an external environment vehicle by one or more sensors of the host vehicle; and

causing a processor of the host vehicle or an electronic control unit of the host vehicle to, in response to the detecting of the traffic object by the one or more sensors, perform the steps of;

establishing a predicted traffic environment of the host vehicle in a time-lateral position domain, the predicted traffic environment comprising an object prediction representation of the traffic object,

determining a start node for the host vehicle, the start node representing a start time and a start lateral position in the predicted traffic environment,

defining an end node for the host vehicle, the end node representing an end time in the predicted traffic environment,

placing several boundary nodes relative to the object prediction representation in the predicted traffic environment between the start node and the end node, wherein the object prediction representation of the traffic object includes a set of points connected by straight lines in the time-lateral position domain, and wherein each boundary node is placed a predetermined lateral distance from a respective corner of the object prediction representation,

setting node connections between the start node, the boundary nodes, and the end node, thereby defining a connectivity map with possible node connections in the predicted traffic environment, and

traversing the start node, the boundary nodes, and the end node using a graph-search algorithm in order to find a path from the start node to the end node with a lowest cost, whereby the path with the lowest cost is selected as an evasive path associated with the host vehicle, wherein the lowest cost is a lowest cost of maximum lateral jerk, a lowest cost of maximum lateral acceleration, or a lowest cost of a sum of lateral movement.

2. The method according to claim 1, wherein said node connections include splines.

3. The method according to claim 1, wherein each spline has a piecewise constant magnitude of lateral jerk.

4. The method according to claim 1, wherein any lateral motion at an end of a node connection is carried over to a start of a subsequent node connection.

5. The method according to claim 1, wherein any node connections passing through the object prediction representation are disregarded when traversing the start node, the boundary nodes, and the end node.

6. The method of claim 1, wherein the predetermined lateral distance is half the width of the host vehicle and an additional predetermined safety margin.

7. The method of claim 6, wherein the predetermined safety margin is in a range of 0-1 m.

8. The method according to claim 1, wherein the placing the several boundary nodes includes giving each boundary node at least one boundary condition.

9. The method of claim 8, wherein the at least one boundary condition is selected from the group comprising:

required lateral velocity, tolerance in lateral position, and tolerance in time.

10. The method according to claim 1, wherein the graph-search algorithm is an A* search algorithm.

11. The method according to claim 1, further comprising:

causing the processor of the host vehicle or the electronic control unit of the host vehicle to transmit a risk assessment output signal based on the selected evasive path.

12. The method of claim 1, further comprising:

transmitting a control signal to an active safety system of the host vehicle based on the selected evasive path, the active safety system including an automatic steering system and/or an automatic braking system, the active safety system configured to cause the host vehicle to automatically move along a particular path, and

causing the active safety system to, in response to the control signal, cause the host vehicle to automatically follow the selected evasive path through the external environment.

13. A non-transitory computer program product comprising code configured to, when executed by a processor or an electronic control unit, perform a method, the method comprising:

establishing a predicted traffic environment of a host vehicle in a time-lateral position domain;

establishing an object prediction representation of a traffic object in the predicted traffic environment based on detection of the traffic object in an external environment by one or more sensors of the host vehicle, the traffic object and the external environment being external to the host vehicle,

determining a start node associated with the host vehicle in the predicted traffic environment, the start node representing a start time and a start lateral position in the predicted traffic environment,

defining an end node associated with the host vehicle, the end node representing an end time in the predicted traffic environment,

establishing a plurality of boundary nodes relative to the object prediction representation in the predicted traffic environment between the start node and the end node, wherein the object prediction representation of the traffic object includes a set of points connected by straight lines in the time-lateral position domain, and wherein each boundary node is placed a particular lateral distance from a respective corner of the object prediction representation;

setting node connections between the start node, the boundary nodes, and the end node, thereby defining a connectivity map with possible node connections in the predicted traffic environment, and

traversing the start node, the boundary nodes, and the end node using a graph-search algorithm in order to find a path from the start node to the end node that is associated with a lowest cost, whereby the path with the lowest cost is selected as an evasive path associated with the host vehicle, wherein the lowest cost is a lowest cost of maximum lateral jerk, a lowest cost of maximum lateral acceleration, or a lowest cost of a sum of lateral movement.

14. The computer program product according to claim 13, wherein the code is stored on a computer readable medium.

15. The non-transitory computer program product of claim 13, the method further comprising:

transmitting a control signal to an active safety system of the host vehicle based on the selected evasive path to cause the active safety system to cause the host vehicle to automatically follow the selected evasive path through the external environment based on the control signal, the active safety system of the host vehicle including an automatic steering system and/or an automatic braking system.

16. A vehicle comprising:

one or more sensors configured to detect a traffic object that is external to the vehicle in an external environment that is external to the vehicle; and

an electronic control unit configured to;

establish a predicted traffic environment of the vehicle in a time-lateral position domain,

establish an object prediction representation of the traffic object in the predicted traffic environment based on the detection of the traffic object by the one or more sensors,

determine a start node associated with the vehicle in the predicted traffic environment, the start node representing a start time and a start lateral position in the predicted traffic environment,

define an end node associated with the vehicle, the end node representing an end time in the predicted traffic environment,

establish a plurality of boundary nodes relative to the object prediction representation in the predicted traffic environment between the start node and the end node, wherein the object prediction representation of the traffic object includes a set of points connected by straight lines in the time-lateral position domain, and wherein each boundary node is placed a particular lateral distance from a respective corner of the object prediction representation;

set node connections between the start node, the boundary nodes, and the end node, thereby defining a connectivity map with possible node connections in the predicted traffic environment, and

traverse the start node, the boundary nodes, and the end node using a graph-search algorithm in order to find a path from the start node to the end node that is associated with a lowest cost, whereby the path with the lowest cost is selected as an evasive path associated with the vehicle, wherein the lowest cost is a lowest cost of maximum lateral jerk, a lowest cost of maximum lateral acceleration, or a lowest cost of a sum of lateral movement.

17. The vehicle of claim 16, further comprising:

an active safety system including an automatic steering system and/or an automatic braking system, the active safety system configured to cause the vehicle to automatically move along a particular path,

wherein the electronic control unit is further configured to transmit a control signal to the active safety system based on the selected evasive path,

wherein the active safety system is further configured to cause the vehicle to automatically follow the selected evasive path through the external environment based on the control signal.