Granted Patent B2

US 12,418,778 · App. 18/550,061 · Granted Sep 16, 2025

Inventors: Liuyang Yang (Vancouver, WA); Ignacio J. Alvarez (Portland, OR); Xiruo Liu (Portland, OR); Kathiravetpillai Sivanesan (Portland, OR); Leonardo Gomes Baltar (Munich, DE)

Assignee: Intel Corporation

H04W4/38H04W4/40G06Q50/40

Abstract

Claims (84)

1. An apparatus, comprising:

memory circuitry to store instructions for operating a detection system of a misbehavior detection service; and

processor circuitry connected to the memory circuitry, wherein the processor circuitry is to execute the instructions to:

receive a Collective Perception Message (CPM) from a source node, wherein the CPM includes a set of perceived objects and one or more free spaces detected by the source node;

determine, based on the received CPM, a source sensor field of view union of the source node (“SFoV union”) and a free space union of the source node (“SFS union”);

perform a first cross check on each perceived object in the set of perceived objects against the SFoV union;

perform a second cross check on the free space included in the received CPM against the SFoV union;

perform a third cross check on each perceived object against the SFS union;

generate a misbehavior report (MR) message including a potential misbehavior of the source node when an inconsistency is declared as a result of at least one cross check including the first cross check, the second cross check, and the third cross check; and

send the MR message to a misbehavior authority.

2. The apparatus of claim 1 , wherein, to perform the first cross check, the processor circuitry is to execute the instructions to:

declare a consistency when an individual perceived object of the set of perceived objects is inside the SFOV union; and

declare the inconsistency when the individual perceived object is outside of the SFoV union.

3. The apparatus of claim 1 , wherein, to perform the second cross check, the processor circuitry is to execute the instructions to:

determine a free space polygon for each source node sensor indicated by the received CPM;

determine whether the free space polygon of each source node sensor is inside the SFOV union;

declare a consistency when a free space polygon of an individual source node sensor is within the SFoV union; and

declare the inconsistency when the free space polygon of the individual source node sensor is outside of the SFoV union.

4. The apparatus of claim 1 , wherein, to perform the third cross check, the processor circuitry is to execute the instructions to:

determine whether each perceived object is outside of the SFS union;

declare a consistency when an individual perceived object of the set of perceived objects is outside the SFS union; and

declare the inconsistency when the individual perceived object is inside of the SFS union.

5. The apparatus of claim 1 , wherein the CPM is a first CPM, and the processor circuitry is to execute the instructions to:

perform a fourth cross check on a speed and an acceleration of a perceived object in the set of perceived objects that is a subject of a second CPM sent by the source node and received before the first CPM; and

declare, based on the fourth cross check, the inconsistency when the speed or the acceleration over a period of time is outside a standard deviation for a corresponding change in position or speed over the period of time.

6. The apparatus of claim 1 , wherein the CPM is a first CPM, and the processor circuitry is to execute the instructions to:

perform a fifth cross check across multiple CPMs received from multiple different source nodes, wherein, to perform the fifth cross check, the processor circuitry is to execute the instructions to:

determine a receiver node (Rx) sensor field of view union (“RxS-FoV union”), wherein the RxS-FoV union is a union of respective FoVs of a set of local sensors at the Rx;

generate or updating an Rx sensor-based environmental map (“RxS-map”) based on the RxS-FoV union such that objects included in the RxS-FoV union are indicated in the RxS-map;

determine, at each time instance when a new CPM is received, where a perceived object included in the new CPM is disposed in the RxS-map; and

declare the inconsistency when the perceived object included in the new CPM is disposed in an area of the RxS-map indicated as being unoccupied.

7. The apparatus of claim 6 , wherein, to perform the fifth cross check, the processor circuitry is to execute the instructions to:

generate or update an Rx CPM-based environmental map (“Rx-CPM-map”) based on data obtained from at least two data sources; and

declare the inconsistency when a first data source indicates that a portion of the Rx-CPM-map is unoccupied and a second data source indicates that the portion of the Rx-CPM-map is occupied.

8. The apparatus of claim 7 , wherein, to generate or update the Rx-CPM-map, the processor circuitry is to execute the instructions to:

label the portion of the Rx-CPM-map as unoccupied when one data source of the at least two data sources indicates the portion is unoccupied and another data source of the at least two data sources indicates that the portion is unknown, wherein the at least two data sources include a set of local sensors of the Rx and individual CPMs of the multiple CPMs;

label the portion of the Rx-CPM-map as occupied when one data source of the at least two data sources indicates the portion is unknown and another data source of the at least two data sources indicates that the portion is occupied; and

generate the Rx-CPM-map using a data fusion algorithm to fuse sensor data obtained from each local sensor of the set of local sensors at the Rx and perceived objects included in the multiple CPMs, and fuse free space indicators included in the multiple CPMs together.

9. The apparatus of claim 8 , wherein the data fusion algorithm is a first data fusion algorithm, and the processor circuitry is to execute the instructions to:

generate the RxS-FoV union using a second data fusion algorithm to only fuse sensor data obtained from each local sensor of the set of local sensors at the Rx together, wherein the first data fusion algorithm is same as the second data fusion algorithm or different than the second data fusion algorithm;

generate the SFoV union using a third data fusion algorithm to fuse sensor data of the source node indicated by the received CPM, wherein the second data fusion algorithm is same as the third data fusion algorithm or different than the third data fusion algorithm; and

generate the SFS union using a fourth data fusion algorithm to fuse free space indications included in the received CPM, wherein the third data fusion algorithm is same as the fourth data fusion algorithm or different than the fourth data fusion algorithm.

10. The apparatus of claim 1 , wherein the apparatus is or includes a robotic system or an Intelligent Transportation System Station (ITS-S), wherein the ITS-S is a vehicle ITS-S (V-ITS-S), a roadside ITS-S(R-ITS-S), or a vulnerable road user (VRU) ITS-S; and the apparatus further comprises: interface circuitry to communicatively couple the apparatus with one or more sensors.

11. A non-transitory computer-readable medium (NTCRM) comprising instructions, wherein execution of the instructions by one or more processors is to cause a a detection system of a misbehavior detection service to:

receive a Collective Perception Message (CPM) from a source node, wherein the CPM includes a set of perceived objects and one or more free spaces detected by the source node;

determine, based on the received CPM, a source sensor field of view union of the source node (“SFoV union”) and a free space union of the source node (“SFS union”);

perform a first cross check on each perceived object in the set of perceived objects against the SFoV union;

perform a second cross check on the free space included in the received CPM against the SFoV union;

perform a third cross check on each perceived object against the SFS union;

generate a misbehavior report (MR) message including a potential misbehavior of the source node when an inconsistency is declared as a result of at least one cross check including the first cross check, the second cross check, and the third cross check; and

send the MR message to a misbehavior authority.

12. The NTCRM of claim 11 , wherein, to perform the first cross check, execution of the instructions is to cause the detection system to:

declare a consistency when an individual perceived object of the set of perceived objects is inside the SFOV union; and

declare the inconsistency when the individual perceived object is outside of the SFOV union.

13. The NTCRM of claim 11 , wherein, to perform the second cross check, execution of the instructions is to cause the detection system to:

determine a free space polygon for each source node sensor indicated by the received CPM;

determine whether the free space polygon of each source node sensor is inside the SFOV union;

declare a consistency when a free space polygon of an individual source node sensor is within the SFOV union; and

declare the inconsistency when the free space polygon of the individual source node sensor is outside of the SFOV union.

14. The NTCRM of claim 11 , wherein, to perform the third cross check, execution of the instructions is to cause the detection system to:

determine whether each perceived object is outside of the SFS union;

declare a consistency when an individual perceived object of the set of perceived objects is outside the SFS union; and

declare the inconsistency when the individual perceived object is inside of the SFS union.

15. The NTCRM of claim 11 , wherein the CPM is a first CPM, and execution of the instructions is to cause the detection system to:

perform a fourth cross check on a speed and an acceleration of a perceived object in the set of perceived objects that is a subject of a second CPM sent by the source node and received before the first CPM; and

declare, based on the fourth cross check, the inconsistency when the speed or the acceleration over a period of time is outside a standard deviation for a corresponding change in position or speed over the period of time.

16. The NTCRM of claim 11 , wherein the CPM is a first CPM, and execution of the instructions is to cause the detection system to:

perform a fifth cross check across multiple CPMs received from multiple different source nodes, wherein, to perform the fifth cross check, the processor circuitry is to execute the program code to:

determine a receiver node (Rx) sensor field of view union (“RxS-FoV union”), wherein the RxS-FoV union is a union of respective FoVs of a set of local sensors at the Rx;

generate or updating an Rx sensor-based environmental map (“RxS-map”) based on the RxS-FoV union such that objects included in the RxS-FoV union are indicated in the RxS-map;

determine, at each time instance when a new CPM is received, where a perceived object included in the new CPM is disposed in the RxS-map; and

declare the inconsistency when the perceived object included in the new CPM is disposed in an area of the RxS-map indicated as being unoccupied.

17. The NTCRM of claim 16 , wherein, to perform the fifth cross check, execution of the instructions is to cause the detection system to:

generate or update an Rx CPM-based environmental map (“Rx-CPM-map”) based on data obtained from at least two data sources; and

declare the inconsistency when a first data source indicates that a portion of the Rx-CPM-map is unoccupied and a second data source indicates that the portion of the Rx-CPM-map is occupied.

18. The NTCRM of claim 17 , wherein, to generate or update the Rx-CPM-map, execution of the instructions is to cause the detection system to:

label the portion of the Rx-CPM-map as unoccupied when one data source of the at least two data sources indicates the portion is unoccupied and another data source of the at least two data sources indicates that the portion is unknown, wherein the at least two data sources include a set of local sensors of the Rx and individual CPMs of the multiple CPMs;

label the portion of the Rx-CPM-map as occupied when one data source of the at least two data sources indicates the portion is unknown and another data source of the at least two data sources indicates that the portion is occupied; and

generate the Rx-CPM-map using a data fusion algorithm to fuse sensor data obtained from each local sensor of the set of local sensors at the Rx and perceived objects included in the multiple CPMs, and fuse free space indicators included in the multiple CPMs together.

19. The NTCRM of claim 18 , wherein the data fusion algorithm is a first data fusion algorithm, and execution of the instructions is to cause the detection system to:

generate the RxS-FoV union using a second data fusion algorithm to only fuse sensor data obtained from each local sensor of the set of local sensors at the Rx together, wherein the first data fusion algorithm is same as the second data fusion algorithm or different than the second data fusion algorithm;

generate the SFoV union using a third data fusion algorithm to fuse sensor data of the source node indicated by the received CPM, wherein the second data fusion algorithm is same as the third data fusion algorithm or different than the third data fusion algorithm; and

generate the SFS union using a fourth data fusion algorithm to fuse free space indications included in the received CPM, wherein the third data fusion algorithm is same as the fourth data fusion algorithm or different than the fourth data fusion algorithm.

20. The NTCRM of claim 11 , wherein the detection system is implemented by a robotic system or an Intelligent Transportation System Station (ITS-S), wherein the ITS-S is a vehicle ITS-S (V-ITS-S), a roadside ITS-S (R-ITS-S), or a vulnerable road user (VRU) ITS-S.

Continuity (2)

References Cited (16)

Cited By (1)

US 12,641,160