Baidu USA LLC

Baidu.com Times Technology (beijing) Co. Ltd.

Jiaxuan Xu

Jiangtao Hu

Jinyun Zhou

Jinghao Miao

Shu Jiang

Qi Luo

Yu Wang

Runxin He

Jiaming Tao

Jingao Wang

Kecheng Xu

Yifei Jiang

Hongyi Sun

Jiacheng Pan

Shiyu Song

Xiangquan Xiao

Yu Long Cao

Chao Ma

Weiman Lin

Kuang Hu

Siyang Yu

Longtao Lin

Zongbao Feng

Wei-Man Lin

In one embodiment, a process is performed during controlling Autonomous Driving Vehicle (ADV). A confidence level associated with a sensed obstacle is determined. If the confidence level is below a confidence threshold, and a distance between the ADV and a potential point of contact with the sensed obstacle is below a distance threshold, then performance of a driving decision is delayed. Otherwise, the driving decision is performed to reduce risk of contact with the sensed obstacle.

Delay decision making for autonomous driving vehicles in response to obstacles based on confidence level and distance

A method of navigating an autonomous driving vehicle (ADV) includes determining a target function for an open space model based on one or more obstacles and map information within a proximity of the ADV, then iteratively performing first and second quadratic programming (QP) optimizations on the target function. Then, generating a second trajectory based on results of the first and second QP optimizations to control the ADV autonomously using the second trajectory. The first QP optimization is based on fixing a first set of variables of the target function. The second QP optimization is based on maximizing a sum of the distances from the ADV to each of the obstacles over a plurality of points of the first trajectory, and minimizing a difference between a target end-state of the ADV and a determined final state of the ADV using the first trajectory.

End dynamics and constraints relaxation algorithm on optimizing an open space trajectory

A simulation of an autonomous driving vehicle (ADV) includes capturing first data that includes a control command output by an autonomous vehicle controller of the ADV, and capturing second data that includes the control command being implemented at a control unit of the ADV. The control command, for example, a steering command, a braking command, or a throttle command, is implemented by the ADV to affect movement of the ADV. A latency model is determined based on comparing the first data with the second data, where the latency model defines time delay and/or amplitude difference between the first data and the second data. The latency model is applied in a virtual driving environment.

Dynamic model with actuation latency

In one embodiment, a computer-implemented method of operating an autonomous driving vehicle (ADV) includes perceiving a driving environment surrounding the ADV based on sensor data obtained from one or more sensors mounted on the ADV, determining a driving scenario, in response to a driving decision based on the driving environment, applying a predetermined machine-learning model to data representing the driving environment and the driving scenario to generate a set of one or more driving parameters, and planning a trajectory to navigate the ADV using the set of the driving parameters according to the driving scenario through the driving environment.

Neural network approach for parameter learning to speed up planning for complex driving scenarios

In one embodiment, a method of training dynamic models for autonomous driving vehicles includes the operations of receiving a first set of training data from a training data source, the first set of training data representing driving statistics for a first set of features; training a dynamic model based on the first set of training data for the first set of features; determining a second set of features as a subset of the first set of features based on evaluating the dynamic model, each of the second set of features representing a feature whose performance score is below a predetermined threshold. The method further includes the following operations for each of the second set of features: retrieving a second set of training data associated with the corresponding feature of the second set of features, and retraining the dynamic model using the second set of training data.

Automated training data extraction method for dynamic models for autonomous driving vehicles

In one embodiment, a computer-implemented method for optimizing a controller of an autonomous driving vehicle (ADV) includes obtaining several samples, each sample having a set of parameters, iteratively performing, until a predetermined condition is satisfied: determining, for each sample, a score according to a configuration of the controller based on the set of parameters of the sample, applying a machine learning model to the samples and corresponding scores to determine a mean function and a variance function, producing a new sample as a minimum of a function of the mean function and the variance function with respect to an input space of the set of parameters, adding the new sample to the several samples, and outputting the new sample as an optimal sample, where parameters of the optimal sample are utilized to configure the controller to autonomously drive the ADV.

Bayesian Global optimization-based parameter tuning for vehicle motion controllers

In one embodiment, a method determines a route from a first location of an autonomous driving vehicle (ADV) to a second location within an open space, the first location being a current location of the ADV. The method determines an objective function based on the route, the objective function having a set of costs for maneuvering the ADV from the first location to the second location. The method determines environmental conditions of the open space and uses the environmental conditions to determine a set of weights, each weight to be applied to a corresponding cost of the objective function. The method optimizes the objective function in view of one or more constraints, such that an output of the objective function reaches minimum while the one or more constraints are satisfied and generates a path trajectory with the optimized objective function to control the ADV autonomously according to the path trajectory.

Open space path planning using inverse reinforcement learning

In one embodiment, method performed by an autonomous driving vehicle (ADV) that determines, within a driving space, a plurality of routes from a current location of the ADV to a desired location. The method determines, for each route of the plurality of routes, an objective function to control the ADV autonomously along the route and, for each of the objective functions, performs Differential Dynamic Programming (DDP) optimization in view of a set of constraints to produce a path trajectory. The method determines whether at least one of the path trajectories satisfies each constraint and, in response to a path trajectory satisfying each of the constraints, selects the path trajectory for navigating the ADV from the current location to the desired location.

Differential dynamic programming (DDP) based planning architecture for autonomous driving vehicles

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.

Autonomous vehicle park-and-go scenario design

In one embodiment, static-state curvature error compensation control logic for autonomous driving vehicles (ADV) receives planning and control data associated with the ADV, including a planned steering angle and a planned speed. A steering command is generated based on a current steering angle and the planned steering angle of the ADV. A throttle command is generated based on the planned speed in view of a current speed of the ADV. A curvature error is calculated based on a difference between the current steering angle and the planned steering angle. The steering command is issued to the ADV while withholding the throttle command, in response to determining that the curvature error is greater than a predetermined curvature threshold, such that the steering angle of the ADV is adjusted in view of the planned steering angle without acceleration.

Static-state curvature error compensation control logic for autonomous driving vehicles

Systems and methods are disclosed for collecting driving data from simulated autonomous driving vehicle (ADV) driving sessions and real-world ADV driving sessions. The driving data is processed to exclude manual (human) driving data and to exclude data corresponding to the ADV being stationary (not driving). Data can further be filtered based on driving direction: forward or reverse driving. Driving data records are time stamped. The driving data can be aligned according to the timestamp, then a standardized set of metrics is generated from the collected, filtered, and time-aligned data. The standardized set of metrics are used to grade the performance the control system of the ADV, and to generate an updated ADV controller, based on the standardized set of metrics.

Cross-platform control profiling tool for autonomous vehicle control

Systems and methods are disclosed for reducing second order dynamics delays in a control subsystem (e.g. throttle, braking, or steering) in an autonomous driving vehicle (ADV). A control input is received from an ADV perception and planning system. The control input is translated in a control command to a control subsystem of the ADV. A reference actuation output is obtained from a storage of the ADV. The reference actuation output is a smoothed output that accounts for second order actuation dynamic delays attributable to the control subsystem actuator. Based on a difference between the control input and the reference actuation output, adaptive gains are determined and applied to the input control signal to reduce error between the control output and the reference actuation output.

Model reference adaptive control algorithm to address the vehicle actuation dynamics

In one embodiment, a system generates a plurality of driving scenarios to train a reinforcement learning (RL) agent and replays each of the driving scenarios to train the RL agent by: applying a RL algorithm to an initial state of a driving scenario to determine a number of control actions from a number of discretized control/action options for the ADV to advance to a number of trajectory states which are based on a number of discretized trajectory state options, determining a reward prediction by the RL algorithm for each of the controls/actions, determining a judgment score for the trajectory states, and updating the RL agent based on the judgment score.

Offline agent using reinforcement learning to speedup trajectory planning for autonomous vehicles

In one embodiment, a computer-implemented method of autonomously parking an autonomous driving vehicle, includes generating environment descriptor data describing a driving environment surrounding the autonomous driving vehicle (ADV), including identifying a parking space and one or more obstacles within a predetermined proximity of the ADV, generating a parking trajectory of the ADV based on the environment descriptor data to autonomously park the ADV into the parking space, including optimizing the parking trajectory in view of the one or more obstacles, segmenting the parking trajectory into one or more trajectory segments based on a vehicle state of the ADV, and controlling the ADV according to the one or more trajectory segments of the parking trajectory to autonomously park the ADV into the parking space without collision with the one or more obstacles.

Segmenting a parking trajectory to control an autonomous driving vehicle to park

In one embodiment, a system uses an actor-critic reinforcement learning model to generate a trajectory for an autonomous driving vehicle (ADV) in an open space. The system perceives an environment surrounding an ADV. The system applies a RL algorithm to an initial state of a planning trajectory based on the perceived environment to determine a plurality of controls for the ADV to advance to a plurality of trajectory states based on map and vehicle control information for the ADV. The system determines a reward prediction by the RL algorithm for each of the plurality of controls in view of a target destination state. The system generates a first trajectory from the trajectory states by maximizing the reward predictions to control the ADV autonomously according to the first trajectory.

Growing community of inventors

Sunnyvale, CA, United States of America

Jiaxuan Xu