Abstract: Method and device for controlling operation of a Self-Driving Car. The method includes determining route-level and lane-level information for generating a graph-structure, applying a first model for assigning costs to respective edges, determining scores for vertices based on the costs, storing the edges with the costs and the vertices with the scores. The method also includes at a given moment in time during operation: acquiring a lane path indicative of a lane segment extending from the current location without a lane-changing manoeuvre. The method includes, for the lane path: identifying a series of vertices covered by the lane segment, applying a second model for assigning additional costs to lane-path-departing edges of the series of vertices thereby determining locally-increased costs, determining a locally-adjusted scores for vertices in the series of vertices based on the locally-increased costs, and identifying a lane path score for the lane path based on the locally-adjusted scores.

Abstract: A method of and processor for determining a trajectory estimation order for vehicles are disclosed. The processor has access to a section of a road map corresponding to surroundings of the one of the vehicles, and road rules associated therewith. The method includes identifying pairs of vehicles that have intersecting potential trajectories and vehicle priority information indicative of priorities of vehicles on the section of the road map. Based on the vehicle priority information, the method includes determining a trajectory estimation order for vehicles on the section of the road map, such that a given vehicle having priority over another vehicle in the vehicle priority information is above another vehicle in the trajectory estimation order.

Abstract: A method and a processor for controlling in-lane movement of a Self-Driving Vehicle (SDV) are provided. The method comprises: acquiring initial kinematic data associated with an obstacle; determining future kinematic data associated with the obstacle; acquiring initial kinematic data associated with the SDV; determining future kinematic data associated with the SDV which is indicative of at least two candidate future states of the SDV at a future moment in time; determining at least two candidate state-transition datasets for the SDV to transition from the initial state of the SDV to a respective one of the at least two candidate future states of the SDV using only in-lane movement; determining, by the electronic device, an energy efficiency score for a respective one of the at least two candidate state-transition datasets; determining a target state-transition dataset for the SDV based on respective energy efficiency scores.

Abstract: A method and an electronic device for merging digital datasets in a single multidimensional space are provided. The method comprises: selecting, by the electronic device, (i) at least some of the plurality of first data points and (ii) at least some of the plurality of second data points; matching, by the electronic device, the first data points with the second data points, thereby determining a plurality of pairs; determining, by the electronic device, a pair-specific error value for a given one of the plurality of pairs; determining, by the electronic device, a weight factor for the given one of the plurality of pairs based on a normal vector associated with the given second data point in the given one of the plurality of pairs; and determining, by the electronic device, a global error value for the second dataset.

Abstract: A method of and system for processing Light Detection and Ranging (LIDAR) point cloud data. The method is executable by an electronic device, communicatively coupled to a LIDAR installed on a vehicle, the LIDAR having a plurality of lasers for capturing LIDAR point cloud data. The method includes receiving a first LIDAR point cloud data captured by the LIDAR; executing a Machine Learning Algorithm (MLA) for: analyzing a first plurality of LIDAR points of the first point cloud data in relation to a response pattern of the plurality of lasers; retrieving a grid representation data of a surrounding area of the vehicle; determining if the first plurality of LIDAR points is associated with a blind spot, the blind spot preventing a detection algorithm of the electronic device to detect presence of at least one object surrounding the vehicle conditioned on the at least one object is present.

Abstract: A method and processor for controlling steering of a Self-Driving Car (SDC) are disclosed. The method includes: acquiring SDC state data associated with vertices in a graph-structure, using the SDC state data for building a polynomial curve, using the polynomial curve for simulating a candidate trajectory of the SDC as if the SDC is steered in accordance therewith and such that a closest allowed steering wheel position to a non-allowed steering wheel position is used instead of the non-allowed steering wheel position, determining an offset between the simulated state and a target state of the SDC, selecting candidate steering profile as a target steering profile of the SDC based on the offset, and using the target steering profile of the SDC for controlling steering of the SDC. A method of training a Machine Learning Algorithm for predicting steering wheel positions to control steering of the SDC is also disclosed.

Abstract: A method and an electronic device for generating a trajectory of a Self-Driving Car (SDC) are provided. The method comprises: determining a presence of at least one third-party object around the SDC; generating a plurality of predicted trajectories for the third-party object, where at least one of the plurality of trajectories includes a maneuver executable, by the third-party object, at a future third-party object location; calculating, for the at least one of the plurality of trajectories including the a respective braking profile associated with the third-party object; in response to the respective braking profile being above a pre-determined threshold, eliminating an associated one of the at least one of the plurality of trajectories from future processing; determining an SDC trajectory based on remaining ones of the plurality of predicted trajectories for the third-party.

Abstract: A method and server for training a machine-learning algorithm (MLA) to detect objects in sensor data acquired by a second sensor mounted on a second vehicle located at a second distance from the objects, the MLA having been trained to detect the objects in sensor data acquired by a first sensor mounted on a first vehicle located at a first distance from the objects. First sensor data acquired by the first sensor on the first vehicle is aligned with second sensor data acquired by the second sensor on the second vehicle. The MLA determines objects and objects classes in the aligned first sensor data. The object classes in the aligned first sensor data are assigned to corresponding portions in the aligned second sensor data. The MLA is trained on the labelled portions in the aligned second sensor data to recognize and classify objects at the second distance.

Abstract: Methods and devices for generating a trajectory for a self-driving car (SDC) are disclosed. The method includes: for a given one of the plurality of trajectory points of a given trajectory: (i) determining a presence of plurality of dynamic objects around the SDC, (ii) applying a first algorithm to determine a set of collision candidates, (iii) generating a segment line for the SDC, (iv) generating a bounding box for each of set of collision candidates, (v) for a given one of the set of collision candidates, determining a distance between the segment line and the respective bounding box to determine a separation distance, (vi) in response to the separation distance being lower than a threshold, determining that the given one of the set of collision candidates would cause the collision with the SDC, (vii) amending at least one of the plurality of trajectory points to render a revised candidate trajectory.

Abstract: A method and an electronic device for determining a presence of an obstacle in a surrounding area of a self-driving car (SDC) are provided. The method comprises receiving sensor data representative of the surrounding area of the SDC in a form of 3D point cloud data; generating, by an MLA, based on the 3D point cloud data, a set of feature vectors representative of the surrounding area; generating, by the MLA, a grid representation of the surrounding area, each given cell of the grid representation including a predicted distance parameter indicative of a distance from the given cell to a closest cell with the obstacle; and using, by the electronic device, the distance parameter to determine presence of the obstacle in the surrounding area of the SDC.

Abstract: A computer-implemented method for processing a 3-D point cloud data and an associated image data to enrich the 3-D point cloud data with relevant portions of the image date. The method comprises generating a 3-D point cloud data tensor representative of information contained in the 3-D point cloud data and generating an image tensor representative of information contained in the image data; and then analyzing the image tensor to identify a relevant data portion of the image information relevant to the at least one object candidate. The method further includes amalgamating the 3-D point cloud data tensor with a relevant portion of the image tensor associated with the relevant data portion of the image information to generate an amalgamated tensor associated with the surrounding area and storing the amalgamated tensor to be used by a machine learning algorithm (MLA) to determine presence of the object in the surrounding area.

Abstract: Methods and devices for generating data for controlling a Self-Driving Car (SDC) are disclosed. The method includes: (i) acquiring a predicted object trajectory for an object, (ii) acquiring a set of anchor points along the lane for the SDC, (iii) for each one of the set of anchor points, determining a series of future moments in time when the SDC is potentially located at the respective one of the set of anchor points, thereby generating a matrix structure including future position-time pairs, (iv) for each future position-time pair in the matrix structure, using the predicted object trajectory for determining a distance between a closest object to the SDC as if the SDC is located at the respective future position-time pair, and (v) storing the distance between the closest object to the SDC in association with the respective future position-time pair in the matrix structure.

Abstract: Methods and devices for determining axis of symmetry of self-driving vehicle (SDV) and for calibrating a Light Detection and Ranging (LIDAR) system are disclosed. One of the axes of the system of coordinates of the LIDAR system extends along a normal direction of a ground surface. the method includes acquiring a subset of detected points in the system of coordinates; generating a subset of mirror-image points based on the subset of detected points; projecting the subset of mirror-image points onto the subset of detected points so as to define pairs of overlapping data points; using symmetrically opposite detected points for determining the axis of symmetry of the SDV in the system of coordinates of the LIDAR system; and calibrating the LIDAR system using an angular offset between the axis of symmetry of the SDV and an other one of the axes of the system of coordinates of the LIDAR system.

Abstract: There is disclosed a method and system for determining a predicted state of a traffic signal. A video of the traffic signal is received. Still images of the traffic signal are generated based on the video. A first machine learning algorithm (MLA) outputs a vector for each bulb in each still image, the vector indicating a predicted status of the bulb. A second MLA determines a predicted state of the traffic signal based on the vectors.

Abstract: Methods and devices for controlling operation of a Self-Driving Car (SDC) are disclosed. The method includes, at a first moment in time during an approach of the SDC to a crosswalk: identifying, an object-inclusion zone in proximity to the crosswalk, determining presence of an object in the object-inclusion zone, determining an interval of time for the object based on movement data of the object, using the interval of time and movement data of the SDC for determining operation-control data for controlling operation of the SDC, and assigning decision data to the object-inclusion zone indicative of a decision.

Abstract: A method for determining a set of dynamic objects in sensor data representative of a surrounding area of a vehicle having sensors, the method being executed by a server, the server executing a machine learning algorithm (MLA). Sensor data is received, and the MLA generates, based on the sensor data, a set of feature vectors. Vehicle data indicative of a localization of the vehicle is received. The MLA generates, based on the set of feature vectors and the vehicle data, a tensor, the tensor including a grid representation of the surrounding area. The MLA generates an mobility mask indicative of grid cells occupied by at least one moving potential object in the grid, and a velocity mask indicative of a velocity associated with the at least one potential object in the grid. The MLA determines, based on the mobility mask and the velocity mask, the set of dynamic objects.

Abstract: Methods and computer devices for determining an angular offset of a radar system having been mounted on a vehicle having a forward direction of travel along a surface. The angular offset is an angle between the scanning direction and the forward direction. The method includes receiving radar data from the radar system. The method includes determining projections of an immobile object velocity in the scanning direction and a direction perpendicular to the scanning direction. The method also includes determining the angular offset of the radar system based on at least one of the projections of the immobile object velocity.