In various examples, a sequential deep neural network (DNN) may be trained using ground truth data generated by correlating (e.g., by cross-sensor fusion) sensor data with image data representative of a sequences of images. In deployment, the sequential DNN may leverage the sensor correlation to compute various predictions using image data alone. The predictions may include velocities, in world space, of objects in fields of view of an ego-vehicle, current and future locations of the objects in image space, and/or a time-to-collision (TTC) between the objects and the ego-vehicle. These predictions may be used as part of a perception system for understanding and reacting to a current physical environment of the ego-vehicle.

In one embodiment, when an autonomous driving vehicle (ADV) is parked, the ADV can determine, based on criteria, whether to operate in an open-space mode or an on-lane mode. The criteria can include whether the ADV is within a threshold distance and threshold heading relative to a vehicle lane. If the criteria are not satisfied, then the ADV can enter the open-space mode. While in the open-space mode, the ADV can maneuver it is within the threshold distance and the threshold heading relative to the vehicle lane. In response to the criteria being satisfied, the ADV can enter and operate in the on-lane mode for the ADV to resume along the vehicle lane.

A set of input conditions is obtained. A plurality of potential decisions is obtained based at least in part on the set of input conditions. A rule-based system is used to process the plurality of potential decisions and obtain a set of one or more updated potential decisions, wherein: the rule-based system specifies a plurality of rules; a rule specifies a rule condition and a corresponding action, wherein when the rule condition is met, the corresponding action is to be performed; and using the rule-based system to process the plurality of potential decisions includes: for a selected potential decision in the plurality of potential decisions, determining whether the rule condition is met for a selected rule among the plurality of rules, wherein the selected rule condition is dependent on, at least in part, the selected potential decision; and in response to the selected rule condition being met, performing the corresponding action. The set of one or more updated potential decisions to be executed is output.

In one embodiment, a method, apparatus, and system for automatically switching on an emergency light at an autonomous driving vehicle (ADV) is disclosed. A present speed of an ADV at a first time instant is determined. A present deceleration of the ADV at the first time instant is determined. Whether the present speed satisfies a present speed condition and whether the present deceleration satisfies a present deceleration condition at the first time instant are determined. In response to determining that the present speed satisfies the present speed condition and that the present deceleration satisfies the present deceleration condition, whether a recent deceleration history of the ADV satisfies a recent deceleration history condition and whether an expected deceleration of the ADV satisfies an expected deceleration condition are determined. If either condition is satisfied, an emergency light of the ADV is automatically switched on.

Embodiments described herein include a method of receiving, by a moving assisting vehicle, a calibration assistance request related to a moving ego vehicle that requested assistance in collaborative calibration of a sensor deployed on the moving ego vehicle. The method further includes analyzing the calibration assistance request to extract at least one of a schedule or an assistance route associated with the requested assistance. The method includes communicating with the moving ego vehicle about a desired location relative to the position of the moving ego vehicle for the moving assisting vehicle to be in order to assist the sensor to acquire information of a target present on the moving assisting vehicle. The method includes facilitating to drive the moving assisting vehicle to reach the desired location to achieve the collaborative calibration of the sensor on the moving ego vehicle.

An automatic driving vehicle includes a touch panel that displays various buttons. An operator can input a drive control instruction in relation to the automatic driving vehicle, with the buttons displayed on the touch panel. In addition to the touch panel, the automatic driving vehicle includes a mechanical door switch for opening and closing an automatic door and a mechanical ramp switch for extending and retracting an automatic ramp, both being disposed spaced apart from the touch panel.

The invention relates to a method for adjusting fully automatic vehicle guidance functions, which are realized by means of a vehicle system of a motor vehicle, during the operation of the motor vehicles in a predefined navigation environment. A stationary infrastructure device that communicates with the motor vehicles is associated with the navigation environment. Function limits of each vehicle guidance function are defined by means of limit operation parameters of the vehicle guidance function. Current traffic situation information describing dynamic objects in the navigation environment is determined by the infrastructure device by means of environment sensors of the navigation environment. The current traffic situation information is used, together with a digital map describing stationary objects and properties of the navigation environment, to determine at least one piece of risk information for each motor vehicle.

Systems and methods are provided for implementing hybrid prediction. Hybrid prediction integrates two deep learning based trajectory prediction approaches: grid-based approaches and graph-based approaches. Hybrid prediction techniques can achieve enhanced performance by combining the grid and graph approaches in a manner that incorporates appropriate inductive biases for different elements of a high-dimensional space. A hybrid prediction framework processor can generate trajectory predictions relating to movement of agents in a surrounding environment based on a prediction model generating using hybrid prediction. Trajectory predictions output from the hybrid prediction framework processor can be used to control an autonomous vehicle. For example, the autonomous vehicle can perform safety-aware and autonomous operations to avoid oncoming objects, based on the trajectory predictions.

A system includes at least one imaging sensor and a processor. The processor is configured to acquire, using the imaging sensor, detected data describing an environment of an autonomous vehicle. The processor is further configured to derive reference data, which describe the environment, from a predefined map, to compute difference data representing a difference between the detected data and the reference data, and to transfer the difference data. Other embodiments are also described.

Current data for a geographic area is accessed. The current data comprises at least one of (a) current traffic data for the geographic area, (b) current incident data for the geographic area, or (c) current weather data for the geographic area. Based on the current data, autonomous driving instructions are determined for the geographic area. A notification comprising the autonomous driving instructions is provided such that the notification is received by a vehicle apparatus located within the geographic area or expected to enter the geographic area based on a route being traversed by a vehicle corresponding to the vehicle apparatus. The vehicle apparatus is onboard the vehicle and is configured to control the vehicle in accordance with the autonomous driving instructions.