Beijing Voyager Technology Co., Ltd.

Didi Research America, LLC

Alok Priyadarshi

Yuzhu Shen

Yun Jiang

Systems and methods are provided for the accurate reproduction during simulation of distributed systems, such as vehicle-based processing systems. In a simulation, the execution times of individual tasks are aligned to those of a recorded process. Data consumed during the recorded process is also consumed during the corresponding portion of the simulation, and data generated during the simulation is associated with timing data from data generated during the recorded processes.

Architecture for distributed system simulation timing alignment

Systems and methods are provided for the realistic simulation of distributed systems, such as vehicle-based processing systems. A statistical property of message timestamps of a plurality of prior messages from one or more recorded processes is determined. During a simulation, messages that are generated are associated with timing data based on the statistical property from the recorded processes.

Architecture for distributed system simulation with realistic timing

A method of communicating messages between modules in a system on a vehicle, each module configured as a publisher node and/or subscriber node, the publisher nodes and the subscriber nodes collectively forming a plurality of nodes that communicate in the operation of the vehicle. One method includes communicating, by a subscriber node, with a registry for information to determine if a new message associated with a first topic is available for reading, determining, by each subscriber node, if a new message associated with the first topic is available for reading, in response to determining a new message associated with the first topic is available for reading, reading from the registry location information indicating where the first message is stored in a first message buffer, and reading, by each subscriber node the first message from the first message buffer using the location information.

Reading messages in a shared memory architecture for a vehicle

Systems and methods are provided for the deterministic simulation of distributed systems, such as vehicle-based processing systems. A distributed system may be represented as a plurality of subsystems or 'nodelets' executing with a single process of a computing device during a simulation. A simulated clock may be used during execution of the nodelets to mitigate the variability in timestamped data that may be caused by latency or jitter. In some embodiments, all timestamps generated during a given frame of work will be assigned the same time value, regardless of when within the frame the timestamps were generated. A task scheduler can update the value of the simulated clock as execution proceeds through different frames of work.

Distributed system task management using a simulated clock

Systems and methods are provided for the deterministic simulation of distributed systems, such as vehicle-based processing systems. A distributed system may be represented as a plurality of subsystems or 'nodelets' executing with a single process of a computing device during a simulation. A task scheduler can schedule the nodelets to execute separately, on a single thread, in serially-occurring frames. In some embodiments, only one nodelet is permitted to execute during any given frame, and therefore only one nodelet is permitted to execute at any given time.

Distributed system execution using a serial timeline

Systems and methods are provided for the configurable simulation of distributed systems, such as vehicle-based processing systems. Simulations may be performed in various modes in which the timing of operations is based on timing observed in real-world instances or other recorded processes. In one simulation mode, the execution times of individual tasks are aligned to those of a recorded process. Data consumed during the recorded process is also consumed during the corresponding portion of the simulation, and data generated during the simulation is associated with timing data from data generated during the recorded processes. In another simulation mode, data generated during the simulation is associated with timing data based on a statistical analysis of timing data from one or more recorded processes. The various simulation modes, and a production mode for real-word instances, may be provided via a unified application programming interface.

Architecture for configurable distributed system simulation timing

Systems and methods are provided for the deterministic simulation of distributed systems, such as vehicle-based processing systems. A distributed system may be represented as a plurality of subsystems or 'nodelets' executing with a single process of a computing device during a simulation. The nodelets may communicate using in-process communication. A task scheduler can schedule the nodelets to execute separately in serially-occurring frames. A simulated clock may be used to mitigate the variability in timestamped data that may be caused by latency or jitter.

Architecture for simulation clock-based simulation of distributed systems

A shared memory message system for communication between modules configured to perform a function related to controlling a vehicle in a computer system on the vehicle. The system may include a plurality of modules each representative as a node, the plurality of modules collectively representative as a plurality of nodes, each node a publisher node and/or a subscriber node, a topic registry having message storage location information, message buffers configured to store published messages, and a communication bus coupled to the topic registry, the message buffers, and the plurality of nodes. The communication bus is configured such that publisher node messages are stored in a message buffer and associated storage location information is stored in the topic registry without going through the kernel of the computer system, and such that subscriber nodes can read messages from a message buffer without going through the computer system kernel.

Shared memory architecture for a vehicle

A method of communicating messages between a plurality of modules in a system on a vehicle, each module of the plurality of modules implemented on at least one processor and configured as a publisher node and/or a subscriber node and collectively forming a plurality of nodes that communicate in the operation of the vehicle. The method may include generating a first message associated with a first topic by a first publisher node, writing, by the first publisher node, the first message in a memory location in a first message buffer of the plurality of message buffers, the first message buffer associated with the first topic and configured to store a plurality of messages associated with the first topic, and writing in a registry information associated with writing the first message, the registry configured to store location information of the first message.

Writing messages in a shared memory architecture for a vehicle

A method of communicating between a plurality of modules on a vehicle, each module configured as a publisher or subscriber node that communicate in the operation of the autonomous vehicle utilizing a shared memory communication system. The method may include generating groups of messages by publisher nodes, each group associated with a unique topic and generated by a single publisher node associated with the unique topic, writing a group of messages in a message buffer associated with a single topic, writing in a registry, location information indicating where the messages were written, reading new message information from the registry, the new message information indicative of whether a new message associated with a particular topic is available, reading location information indicating where the new message is stored if a new message is available, and reading the new message from the respective message buffer.

Growing community of inventors

Mountain View, CA, United States of America

Alok Priyadarshi