An autonomous driving control apparatus may include an image capturing unit configured to capture a driving road image of an ego vehicle; and a control unit configured to generate virtual lanes on a road with no lines, the virtual lanes corresponding to the number of lanes which are decided based on the width of the road with no lines, when the road with no lines is detected in front of the ego vehicle through the driving road image captured by the image capturing unit, and control autonomous driving of the ego vehicle to drive on one of the generated virtual lanes. When a front object is detected from the driving road image, the control unit may control the autonomous driving of the ego vehicle to avoid the front object and drive on the road with no lines.

A vehicle safety system within an autonomous or semi-autonomous vehicle may predict and avoid collisions between the vehicle and other moving objects in the environment. The vehicle safety system may determine one or more perturbed trajectories for another object in the environment, for example, by perturbing the state parameters of a perceived trajectory associated with the object. Each perturbed trajectory may be evaluated to determine whether it intersects or potentially collides the planned trajectory of the vehicle. In some examples, the vehicle safety system may aggregate the results of analyses of multiple perturbed trajectories to determine a collision probability and/or additional weights or adjustment factors associated with the collision prediction, and may determine actions for the vehicle to take based on the collision predictions and probabilities.

The technology relates to articulated autonomous vehicles that can potentially jackknife. To avoid or mitigate such hazardous conditions, the current state of the vehicle is evaluated against the vehicle's planned trajectory, for instance as it drives along a freeway or surface streets. When the evaluation indicates a likelihood of jackknifing, an automated braking approach is implemented using elective braking to stabilize the vehicle. The braking approach can depend on whether the situation involves tractor jackknifing or trailer jackknifing, and one or more different braking mechanisms can be employed for a selective modulation of the braking profile to address actual jackknifing or to prevent the vehicle from entering a jackknifing situation.

Various embodiments include methods and vehicles, such as a semi-autonomous vehicle, for safely operating a vehicle based on vehicle-control gestures made by an occupant. Exemplary implementations may include determining a first vehicle action by applying a first passenger profile to a detected first vehicle-control gesture performed by a first passenger, determining whether the first vehicle action is safe to perform, and operating the vehicle to implement the first vehicle action in response to determining that the first vehicle action is safe for the vehicle and occupants. The first passenger profile may be selected from a plurality of passenger profiles to normalize vehicle-control gestures received from the first passenger. The vehicle control gesture may be ignored or a vehicle action differing from but similar to the first vehicle action may be implemented in response to determining that the first vehicle action is unsafe for the vehicle or occupants.

A parking assist system with improved avoidance steering control and a method thereof are provided. The parking assist system includes a sensor fusion calculation module that fuses sensing data of an ultrasonic sensor and image data of an SVM camera to calculate position information of an object with respect to a current location of a vehicle and a parking assist module that avoids the object based on the position information and performs steering control of the vehicle for autonomous parking. The parking assist system accurately detects the object on a position departing from a field of view (FOV) of the ultrasonic sensor and performs avoidance steering robust to a surrounding object upon autonomous parking.

A method and system for detecting a collision by an autonomous robot based on a multi-sensor long short-term memory (LSTM) are disclosed. The method includes generating an input of an LSTM model by combining an input image received from an autonomous robot, light detection and ranging (LiDAR) distance information, and acceleration information, learning a collision alert situation by inputting the input to the LSTM model, and determining a collision situation using an output of the LSTM model and a fully connected neural network (FNN) model.

A vehicle driving control system includes: a first road-traffic detection information acquisition unit and a first communicator of a mobile body; a road traffic information recognition unit, a control information calculation unit, and a second communicator of a traffic control apparatus; a second road-traffic detection information acquisition unit and a third communicator of a monitoring apparatus; and a driving control execution unit of a vehicle. The first and second road-traffic detection information acquisition units respectively acquire first and second road-traffic detection information. The road traffic information recognition unit recognizes road traffic information from the first and second road-traffic detection information received by the second communicator respectively through the first and third communicators. The control information calculation unit calculates, from the road traffic information, control information for the vehicle. The driving control execution unit controls driving based on the control information received by the first communicator through the second communicator.

A driver management system and a method of operating the same. The driver management system includes: a monitoring part determining a level of driving attention of a driver; and an autonomous driving part receiving the level of driving attention from the monitoring part, determining a level of necessary attention required according to a required level of autonomous driving, and controlling an autonomous driving function of the vehicle according to the required level of autonomous driving when the level of driving attention satisfies the level of necessary attention.

A computer includes a processor and a memory storing instructions executable by the processor to receive operating data including occupant data indicating one or more occupant actions to operate a vehicle and location data specifying a location of a vehicle during collection of the occupant data, assign the operating data to a respective set in a plurality of sets, each set including operating data collected during a different specified period of time from other sets, input the plurality of sets to a machine learning program trained to output an identification of a hazard at a specified location, the hazard being a roadway condition or an obstacle that changes operation of the vehicle from a default operation, output the hazard at the specified location from the machine learning program, and send a message to a vehicle including the hazard at the specified location.

A vehicle navigation system may comprise a memory including instructions and circuitry configured by the instructions to identify a target vehicle in an environment of a vehicle that includes the vehicle navigation system. The circuitry may receive image data of the target vehicle from at least one image capture device of the vehicle; identify, based on analysis of the image data, one or more situational characteristics of the target vehicle, the situational characteristics of the target including an indication that the target vehicle has an activated blinker; and change a navigational state of the vehicle to allow an action of the target vehicle. The vehicle may be configured to cause the change in the navigational state based on a determination that the one or more situational characteristics, including the activated blinker, indicate that the target vehicle would benefit from the change in the navigational state.