Cortica Ltd.

Autobrains Technologies Ltd.

Cartica Ai Ltd.

Other

Cartica Al Ltd.

Igal Raichelgauz

Karina Odinaev

Yehoshua Y Zeevi

Adam Harel

Boaz Mell

Karina Ordinaev

Roi Saida

Eli Passov

Isaac Misri

Armin Biess

Julius Engelsoy

Kirill Dyagilev

Adrian Kaho Chan

Omer Jackobson

Alexander Dikman

Shay Ieizerovitch

Tomer Cohen

Lior Eldar

Tomer Livne

Idan Geller

Ruth Bernstein

A method of automatically identifying faulty representations across far-near classification detections, the method includes continuously obtaining, across a specified period of time, classification decisions for a road element, where the road element is trackable by using at least one sensing unit across the specified period of time and across a range of distances from the at least one sensing unit, automatically searching, by a machine learning process, for a discrepancy between classification decisions made at different distances of the range of distances; determining in real time, based on an outcome of the automatically searching whether the detections made with respect to the road element are faulty; and providing a faulty representation indication with respect to the classification decisions for the road element.

Automatically identifying faulty representations across classification detections

A method of producing a group of neural networks, the method includes determining a first layer potion that is shared by a first neural network sub-group of the group; determining second layer portions that are sharable by second neural network sub-groups, such that different second layer portions are shared by different second neural network sub-groups of the group; and determining third layer portions that are sharable by third neural network sub-groups, such that different third layer portions are shared by different third neural network sub-groups of the group; wherein each neural network of the group further comprises a unique combination of layer portions.

Shared representation of neural network resources

A method for A method for automatic emergency braking (AEB) of a vehicle, the method may include obtaining sensed information regarding an environment of the vehicle and regarding a motion of the vehicle; determining an occurrence of current situation, based on the sensed information; selecting, based on the current situation, a selected narrow AI agent out of multiple narrow AI agents; wherein different narrow AI agents are trained, using a machine learning process, to perform AEB related decisions at different scenarios; wherein the selected narrow AI agent is associated with the occurrence of the current situation; processing, by the selected narrow AI agent, at least one out of (a) at least a first part of the sensed information, and (b) an outcome of a pre-processing of at least a second part of the sensed information; wherein the processing results in a AEB related decision; and responding to the AEB related decision, wherein the responding comprises at least one out of (a) executing the AEB related decision, and (b) suggesting executing the AEB related decision.

Ensemble of narrow AI agents for autonomous emergency breaking

A method that is computer implemented and is for solving inaccuracies associated with object detection, the method includes automatically evaluating, by a controller, an accuracy of signatures for use in the object detection, the signatures were generated by an adaptable artificial intelligence (AI) system. When finding an erroneous signature of the signatures, by the controller, triggering a generation of (a) a narrow AI agent configured to solve an error associated with the erroneous signature and (b) a router that routes a sensed information unit (SIU) associated with the erroneous signature to the narrow AI agent, wherein the erroneous signature, when used for the object detection, results in an object detection error.

Solving inaccuracies associated with object detection

A method that is automatically validating of signatures for object detection, the method includes automatically evaluating, by a controller, a signature being generated by an adaptable artificial intelligence (AI) system and stored in a memory, by determining whether the signature introduced an error in a previously accurate detection of an object; and triggering a response to address an outcome of the evaluating with respect to a problem found during the automatically evaluating.

Automatically validating of signatures for object detection

A method for lane boundary detection. The method may include obtaining an image of an environment of a vehicle. The environment may include at least one lane boundary; the images may include image segments; calculating image segment lane boundary metadata for relevant image segments. The calculating is executed by a neural network and is based on the image segments; wherein each relevant image segment comprises one or more lane boundary points candidates; wherein image segment lane boundary metadata that is related to a relevant image segment which may include (i) location information about location of the one or more lane boundary points candidates of the relevant image segment, and (ii) angular information regarding an angle between lane boundary points candidates of the image; and determining an estimate of a lane boundary of the at least one lane boundary; wherein the determining is based on the image segment lane boundary metadata of the relevant image segments; wherein the determining comprises applying a process that differs from neural network processing.

Lane boundary detection

A method for safe transfer between manned and autonomous driving modes, the method may include detecting, based on first sensed information sensed during a first period, a situation related to an environment of the autonomous vehicle; searching for one or more matching concepts of a group of reference concepts, to which the situation belongs, each reference concept of the group represents a plurality of situations and has a reference concept safety level; wherein for each reference concept of at least a sub-group of the group of reference concepts the safety level of the reference concept is based on a tested success level of at only some of the plurality of scenarios represented by the reference concept; and determining, based on an outcome of the searching, whether the vehicle is capable to safely autonomously drive through the environment.

Safe transfer between manned and autonomous driving modes

A method for bump alert, the method may include receiving by a vehicle computerized system, at least one visual bump indicator that is visible before driving over at least one bump; obtaining sensed information regarding an environment of the vehicle; processing the sensed information, wherein the processing comprises searching a visual bump indicator of the at least one visual bump indicator; determining whether the vehicle approaches a bump; and generating the bump alert when determining that the vehicle approaches the bump.

Bump alert

A method for off road driving, the method may include obtaining environment sensed information about an environment of a vehicle of a certain model, by one of more vehicle sensors of the vehicle and while driving over an off road path; detecting, by a machine learning process, an off road driving event; determining, by the machine learning process, a characteristic behavior of vehicles of the certain model when facing the off road driving event; and responding, at least in part by the machine learning process, to the occurrence of the off road driving event.

Off road assistance

A method for communicating with a driver of a vehicle, the method includes (i) obtaining object information regarding one or more objects located within an environment of the vehicle; (ii) analyzing the object information; (iii) determining, by using one or more neural network (NNs), and based on the object information, a virtual force associated with a physical model and representing an impact of the one or more objects on a behavior of the vehicle; and (iv) determining, based on at least the virtual force, a force feedback for use in providing a physical restraining force to be applied in association with a driver action made by the driver using the vehicle.

Virtual fields driving related operations

A method for bypass alert, the method includes (i) obtaining sensed information about an environment of the vehicle, by one of more vehicle sensors of a vehicle. The environment of the vehicle includes a passing lane and a current lane; (ii) determining whether a driver initiates a bypass of another vehicle to be bypassed, by a first machine learning process (MLP), based on at least one out of vehicle parameters and the sensed information (iii) determining that the vehicle entered the passing lane; (iv) determining, by a second (MLP), a determined time to collision (TTC) to an incoming vehicle that is located at the passing lane and drives towards the vehicle; (v) determining, based on at least on the determined TTC and an allowable TTC to provide a safety parameter of the bypass; and (vi) responding to the safety parameter.

Bypass assistance

A method for providing an in-vehicle indication, the method may include obtaining, during a driving session of a vehicle, external audio information about audio generated outside the vehicle; determining whether the audio generated outside the vehicle comprises driving affecting audio; and generating the in-vehicle indication, within the vehicle, when determining that the audio generated outside comprises driving affecting audio. The in-vehicle indication is indicative about the driving affecting audio generated outside the vehicle.

Alerting on driving affecting signal

A method for intersection assistance, the method may include obtaining sensed information regarding an environment of the vehicle; determining an occurrence of an intersection related situation, based on the sensed information; generating one or more intersection driving related decisions; wherein the generating comprises processing, by one or more narrow AI agents of a group of narrow AI agents, at least one out of (a) at least a first part of the sensed information, and (b) an outcome of a pre-processing of at least a second part of the sensed information; and responding to the one or more intersection driving related decisions; wherein the responding comprises at least one out of (a) executing the one or more intersection driving related decisions, and (b) suggesting executing the one or more intersection driving related decisions.

Ensemble of narrow AI agents for intersection assistance

A method for real time management of detected issues, the method includes producing, by a classification unit having a neural network, a classification decision for sensed information obtained in an environment of a vehicle; generating, by one or more computing devices with auto-labeling capabilities, an automated ground truth labeling for the sensed information; detecting, by the one or more computing devices and based on a performance indication related to the automated ground truth labeling, an issue with respect to the classification decision; and responsive to the detecting, addressing the detected issue in a driving in the environment of the vehicle by a computer device associated with the vehicle, using a signature generated in association with at least the classification decision or with the detected issue. The neural network is in a same state in the producing of the classification decision, the detecting the issue, and the addressing the detected issue.

Real time management of detected issues

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.

Growing community of inventors

Tel Aviv, Israel

Igal Raichelgauz