Multi-sensor probabilistic object detection and automated braking
First Claim
1. A method comprising performing, by a vehicle controller:
- performing a plurality of iterations by;
receiving a plurality of sensor outputs each including an obstruction-indicating feature from a plurality of sensors having a plurality of different sensing modalities;
for a first iteration of the plurality of iterations, calculating a probability according to each sensor output of the plurality of sensor outputs and an estimated distance to the obstruction-indicating feature in the each sensor output, the probability indicating a probability of obstruction;
for subsequent iterations of the plurality of iterations, updating the probability according to each sensor output of the plurality of sensor outputs and the estimated distance to the obstruction-indicating feature in the each sensor output; and
when the probability is sufficiently high to perform collision avoidance, performing collision avoidance;
wherein at least one sensor of the plurality of sensors has multiple directional variances associated therewith;
wherein updating the probability comprises initiating for the first iteration of the plurality of iterations and updating for the subsequent iterations of the plurality of iterations each probability of a plurality of directional probabilities according to each sensor output of the at least one sensor according to both of the estimated distance and a directional variance of the directional variances corresponding to the each probability.
2 Assignments
0 Petitions
Accused Products
Abstract
A controller receives outputs from a plurality of sensors such as a camera, LIDAR sensor, RADAR sensor, and ultrasound sensor, which may be rearward facing. A probability is updated each time a feature in a sensor output indicates presence of an object. The probability may be updated as a function of a variance of the sensor providing the output and a distance to the feature. Where the variance of a sensor is directional, directional probabilities may be updated according to these variances and the distance to the feature. If the probability meets a threshold condition, actions may be taken such as a perceptible alert or automatic braking. The probability may be decayed in the absence of detection of objects. Increasing or decreasing trends in the probability may be amplified by further increasing or decreasing the probability.
23 Citations
16 Claims
-
1. A method comprising performing, by a vehicle controller:
-
performing a plurality of iterations by; receiving a plurality of sensor outputs each including an obstruction-indicating feature from a plurality of sensors having a plurality of different sensing modalities; for a first iteration of the plurality of iterations, calculating a probability according to each sensor output of the plurality of sensor outputs and an estimated distance to the obstruction-indicating feature in the each sensor output, the probability indicating a probability of obstruction; for subsequent iterations of the plurality of iterations, updating the probability according to each sensor output of the plurality of sensor outputs and the estimated distance to the obstruction-indicating feature in the each sensor output; and when the probability is sufficiently high to perform collision avoidance, performing collision avoidance; wherein at least one sensor of the plurality of sensors has multiple directional variances associated therewith; wherein updating the probability comprises initiating for the first iteration of the plurality of iterations and updating for the subsequent iterations of the plurality of iterations each probability of a plurality of directional probabilities according to each sensor output of the at least one sensor according to both of the estimated distance and a directional variance of the directional variances corresponding to the each probability. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A system comprising:
-
a plurality of sensors having a plurality of sensing modalities; a vehicle controller operably coupled to the plurality of sensors, the vehicle controller programmed to— perform a plurality of iterations each including; receiving a plurality of sensor outputs each including an obstruction-indicating feature indicating obstruction, the plurality of sensors having a plurality of different sensing modalities; for a first iteration of the plurality of iterations, calculating a probability according to each sensor output of the plurality of sensor outputs and an estimated distance to the obstruction-indicating feature in the each sensor output, the probability indicating a probability of obstruction; for iterations subsequent to the first iteration updating the probability according to each sensor output of the plurality of sensor outputs and the estimated distance to the obstruction-indicating feature in the each sensor output; reducing the probability if a period of time occurs in which no obstruction-indicating features are detected in outputs of the plurality of sensors; if a decreasing trend in the probability is detected, further reduce the probability in response to the decreasing trend; if an increasing trend in the probability is detected, additionally increase the probability in response to the increasing trend; and if the probability meets a threshold condition, perform collision avoidance. - View Dependent Claims (11, 12, 14, 15, 16)
-
-
13. A system comprising:
-
a plurality of sensors having a plurality of sensing modalities; a vehicle controller operably coupled to the plurality of sensors, the vehicle controller programmed to— perform a plurality of iterations each including; receiving a plurality of sensor outputs each including an obstruction-indicating feature indicating obstruction, the plurality of sensors having a plurality of different sensing modalities; for a first iteration of the plurality of iterations, calculating a probability according to each sensor output of the plurality of sensor outputs and an estimated distance to the obstruction-indicating feature in the each sensor output, the probability indicating a probability of obstruction; for iterations subsequent to the first iteration updating the probability according to each sensor output of the plurality of sensor outputs and the estimated distance to the obstruction-indicating feature in the each sensor output and if the probability meets a threshold condition, perform collision avoidance wherein at least one sensor of the plurality of sensors has multiple directional variances associated therewith; and wherein the vehicle controller is further programmed to initiate for a first iteration of the plurality of iterations and update for subsequent iterations of the plurality of iterations each probability of a plurality of directional probabilities according to each sensor output of the at least one sensor according to both of the estimated distance and a directional variance of the directional variances corresponding to the each probability.
-
Specification