Prediction of pedestrian entering the drivable space
A method for unregulated pedestrian step down (PSD) alert, the method may include (i) receiving, by a vehicle computerized system, a unregulated PSD indicators that are indicative of unregulated PSD situations; (ii) obtaining sensed information regarding an environment of the vehicle; (iii) processing the sensed information, wherein the processing comprises searching for at least one of the unregulated PSD indicators; and (iv) autonomously determining, when finding the at least one of the unregulated PSD identifiers, that the vehicle is approaching a situation in which a pedestrian is expected to step down into a drivable space within the environment of the vehicle. Wherein the autonomously determining triggering a response to the finding, the response is an immediate response and is executed by at least the vehicle computerized system.
1. A method for unregulated pedestrian step down alert, the method comprises:
receiving, by a vehicle computerized system that comprises one or more integrated circuits, unregulated pedestrian step down (PSD) indicators that are indicative of unregulated PSD situations;
obtaining sensed information regarding an environment of a vehicle;
processing, by the vehicle computerized system, the sensed information, wherein the processing comprises searching for at least one of the unregulated PSD indicators; and
autonomously determining by the vehicle computerized system, when finding the at least one of the unregulated PSD identifiers, that the vehicle is approaching a situation in which a pedestrian is expected to step down into a drivable space within the environment of the vehicle; and
responding to the determining by autonomously driving the vehicle by an autonomous driving system of the vehicle;
wherein the unregulated PSD indicators are generated, using a neural network, by:
obtaining a first group of untagged video recordings;
identifying, within the first group of untagged video recordings, unregulated PSD events;
generating the unregulated PSD indicators by a generation process that comprises processing PSD event video segments, wherein each PSD event video segment starts at a period of time preceding a starting of a corresponding unregulated PSD event of the unregulated PSD events; and
removing identifiers of objects that appear in the unregulated PSD event video segments and are not a characteristic of the unregulated PSD events.
2. The method according to claim 1 , wherein the removing comprising searching for objects that appear, above a predefined probability, in video segments that differ from the unregulated PSD event video segments.
3. The method according to claim 1 , wherein the generating of the unregulated PSD indicators comprises:
generating unregulated PSD event video segments signatures;
clustering the unregulated PSD event video segments signatures to provide multiple clusters; and
generating cluster signatures for the multiple clusters.
4. The method according to claim 3 wherein at least some of the unregulated PSD indicators are cluster signatures.
5. The method according to claim 1 , wherein the generation process comprises classifying the pedestrians to multiple pedestrian classes.
6. The method according to claim 5 comprising associating the pedestrian classes with unregulated PSD indicators that are relevant to the pedestrian classes.
7. The method according to claim 1 , further comprising associating probability of occurrence information to the unregulated PSD indicators.
8. The method according to claim 1 , wherein the responding comprises transferring control over the vehicle from a human driver of the vehicle to the autonomous driving system.
9. A non-transitory computer readable medium for unregulated pedestrian step down alert, the non-transitory computer readable medium stores instructions for:
receiving, by a vehicle computerized system that comprises one or more integrated circuits, unregulated pedestrian step down (PSD) indicators indicative of unregulated PSD situations;
obtaining sensed information regarding an environment of a vehicle;
processing the sensed information, wherein the processing comprises searching for at least one of the unregulated PSD indicators;
autonomously determining, by the vehicle computerized system, when finding the at least one of the unregulated PSD identifiers, that the vehicle is approaching a situation in which a pedestrian is expected to step down into a drivable space within the environment of the vehicle;
responding to the determining by autonomously driving the vehicle by an autonomous driving system of the vehicle;
wherein the unregulated PSD indicators are generated by:
obtaining a first group of untagged video recordings;
finding, within the first group of untagged video recordings, unregulated PSD events;
generating the unregulated PSD indicators by a generation process that comprises processing PSD event video segments, wherein each PSD event video segment starts at a period of time that precedes a starting of a corresponding unregulated PSD event of the unregulated PSD events; and
removing identifiers of objects that appear in the unregulated PSD event video segments and that are not a characteristic of the unregulated PSD events.
10. The non-transitory computer readable medium according to claim 9 , wherein the removing comprising searching for objects that appear, above a predefined probability, in video segments that differ from the unregulated PSD event video segments.
11. The non-transitory computer readable medium according to claim 9 , wherein the generating of the unregulated PSD indicators comprises:
generating unregulated PSD event video segments signatures;
clustering the unregulated PSD event video segments signatures to provide multiple clusters; and
generating cluster signatures for the multiple clusters.
12. The non-transitory computer readable medium according to claim 11 wherein at least some of the unregulated PSD indicators are cluster signatures.
13. The non-transitory computer readable medium according to claim 9 , wherein the generation process comprises classifying the pedestrians to multiple pedestrian classes.
14. The non-transitory computer readable medium according to claim 13 that stores instructions for associating the pedestrian classes with unregulated PSD indicators that are relevant to the pedestrian classes.
15. The non-transitory computer readable medium according to claim 9 that stores instructions for associating probability of occurrence information to the unregulated PSD indicators.
16. The non-transitory computer readable medium according to claim 9 , further storing instructions for transferring control over the vehicle from a human driver of the vehicle to the autonomous driving system.