APOLLO AUTONOMOUS DRIVING USA LLC

A printed circuit board (PCB) includes one or more signal layers in between a first surface and a second surface, the one or more signal layers including a first signal layer. The PCB includes a connector interface disposed on the first surface and a first signal via electrically coupled to the connector interface. The PCB includes first transmission line disposed at the first signal layer electrically coupling a first location to a second location of the first signal layer and a second signal via disposed at the second location. The PCB includes a power-over-cable circuit disposed on the first surface and electrically coupled to the second signal via and a receiver circuit disposed on the first surface adjacent to the power-over-cable circuit. The PCB includes a second transmission line disposed on the first surface layer electrically coupling the power-over-cable circuit to the receiver circuit.

Smooth and agile acceleration of an autonomous driving vehicle (ADV) that is being held at a standstill or is otherwise at a stop is important for ADV safety, efficiency, and occupant enjoyment. For example, as a practical matter, an ADV at an uncontrolled intersection should quickly and smoothly drive-off when it is its turn to advance—otherwise, it may get stuck at the intersection. Embodiments herein provide systems and methods to facilitate a pleasant and comfortable ride while having the ADV drive off smoothly (e.g., not a large jerk) and safely (e.g., not slipping) during driving from a stop—regardless of whether the ADV is on a slope or on a neutral (i.e., not sloped) surface. In one or more embodiments, the drive-off process may be divided into several stages from a standstill to an acceleration settle stage.

A vehicle operating system (VOS) in an autonomous driving vehicle (ADV) can communicate with a cloud platform to automatically train AI models. The VOS collects real-time data from the ADV, and generates inference data based on the real-time data using a teacher edge model of an AI model and generates second inference data based on the real-time data using a student edge model of the AI model. The VOS then obtains one or more differences between the first inference data and the second inference data, and retrains the student edge model of the AI model based on the one or more differences. Both real-time data and the retrained student edge model are uploaded to a cloud platform for use in upgrading the student edge model and the teacher edge model on the cloud platform. The upgraded teacher edge model and the student edge model can be redeployed over-the-air (OTA) through a software define process. The above process of training AI models can be repeated in a closed-loop automatically without user intervention.

Operating a vehicle with an open door creates risks. Embodiments herein consider both when a vehicle is in an autonomous driving (AD) mode and a manual mode. When in AD mode, if an occupant opens the doors while the vehicle is moving, the control system may lower the speed to stop and may park the vehicle (or not park it depending upon the control policy). If an occupant opens the doors when the vehicle is stationary, the control system may keep the vehicle from rolling and from brake system failures. When the vehicle is in a manual model and a door is opened when the vehicle is moving, a warning message may be triggered; whereas, if a door is opened when the vehicle is stationary, the control system may keep the vehicle from rolling and from brake system failures. Warning messages may be triggered for different stages/conditions.

Embodiments presented herein include systems, methods, and non-transitory computer-readable medium or media for controlling power for a vehicle that utilizes one or more computing systems for control, such as an autonomous vehicle. Embodiments comprise multi-tiered distributed power supplies. In one or more embodiments, the overall power system may be structured into distinct levels for enhanced reliability and efficiency. Certain power systems may be tasked with supporting certain systems of the vehicle, which may be correlated to the amount of power (i.e., maximum power and duration) the tier can supply. Because of the inherent risks related to vehicles, a paramount emphasis of the power system design and operation is related to safety. In one or more embodiments, a safety power supply is included as a backup system and may be used to perform one or more safety shutdown procedures.

A system generates a first variant of a first machine learning (ML) model, the first variant associated with an initial hyperparameter value. The system determines a prediction metric for the first variant of the first ML model, the prediction metric indicating an accuracy of the behavior prediction. The system generates an estimated simulation metric for the first variant of the first ML model by applying a second ML model to the prediction metric. The system identifies a first hyperparameter associated with a second variant of the first ML model, the second variant of the first ML model having a corresponding prediction metric and a corresponding estimated simulation metric that meet a first predetermined criteria, the second variant of the first ML model is used by an autonomous driving vehicle (ADV) for behavior prediction.

In one embodiment, a power circuit for an autonomous driving vehicle (ADV) includes a first power source having a first power rating. The circuit includes a second power source having a second power rating. The circuit includes a switch coupled to the first and second power sources, the switch being configured to determining a power operating mode of an accelerator device, where the power operating mode is one of a primary mode, a secondary mode, and a dual mode. The circuit includes a first power transistor coupled between the first power source and the accelerator device and a second power transistor coupled between the second power source and the accelerator device, where the first and second power transistors are operable according to the power operating mode.

This disclosure provides systems and methods for deceleration control during emergency braking using control algorithm override. An embodiment of the present disclosure provides a computer-implemented method for deceleration by a controlling device of an autonomous driving vehicle (ADV). The method includes engaging a braking system of the ADV to decelerate the ADV using a first deceleration algorithm. When an onset of discrepancy between a velocity of the ADV and a corresponding wheel speed of the ADV is detected, a processing device, based on the discrepancy, overrides an output of the first deceleration algorithm using a second deceleration algorithm that prioritizes in reducing the discrepancy between the velocity of the ADV and the corresponding wheel speed of the ADV over a target rate of deceleration computed by the first deceleration algorithm.

Embodiments herein relate to robust methodologies for autonomous parking. In one or more embodiments, an autonomous vehicle may determine the slope of a road based upon one or more inputs and a pre-defined slope definition and may also determine curb/no curb status of a parking location. Given the determined road conditions, such as slope and no curb/curb, embodiments determine wheel direction and angle that the vehicle should achieve to properly park. Embodiments also include countermeasures if one or more issues prohibit the vehicle from achieving a desired parking condition.

Embodiments herein comprise monitoring factors related to braking of an autonomous vehicle. In response to determining, based upon data from one or more sensors, whether a failure state from a set of one or more failure states has occurred, a failure log is updated to record the occurrence of the detected failure state. In one or more embodiments, the failure states have associated severity levels, and the updated log and associated severities are used to determine a braking safety or failure metric for the vehicle. In one or more embodiments, if the vehicle has a metric corresponding to a rule that correlates the braking safety metric value to a set of one or more actions, the corresponding action(s) may then be implemented. For example, if the braking safety metric has a value corresponding to a rule that indicates the vehicle is unsafe, the vehicle is parked.