Abstract: Aspects of the present disclosure provide techniques for training a machine-learning model to compress map data for use online by an autonomous vehicle and techniques for compressing map data using the trained machine-learning model. A system includes a computing device configured to deploy a simulation environment initiating an instance of a virtual vehicle, execute iterations of a simulation of the virtual vehicle, wherein each iteration: deploys a set of map data compressed by the machine-learning compression model and causes the virtual vehicle to execute control operations based on the deployed set of map data, evaluate performance of the executed control operations by the virtual vehicle based on the compressed map data for each iteration, and train the machine-learning compression model to compress map data such that the evaluated performance of the executed control operations by the virtual vehicle exceeds a performance threshold.

Abstract: Aspects of the present disclosure provide techniques for training a machine-learning model to compress map data for use online by an autonomous vehicle and techniques for compressing map data using the trained machine-learning model. A system includes a computing device configured to deploy a simulation environment initiating an instance of a virtual vehicle, execute iterations of a simulation of the virtual vehicle, wherein each iteration: deploys a set of map data compressed by the machine-learning compression model and causes the virtual vehicle to execute control operations based on the deployed set of map data, evaluate performance of the executed control operations by the virtual vehicle based on the compressed map data for each iteration, and train the machine-learning compression model to compress map data such that the evaluated performance of the executed control operations by the virtual vehicle exceeds a performance threshold.

Abstract: Aspects of the present disclosure provide techniques for training a machine-learning model to compress map data for use online by an autonomous vehicle and techniques for compressing map data using the trained machine-learning model. A system includes a computing device configured to deploy a simulation environment initiating an instance of a virtual vehicle, execute iterations of a simulation of the virtual vehicle, wherein each iteration: deploys a set of map data compressed by the machine-learning compression model and causes the virtual vehicle to execute control operations based on the deployed set of map data, evaluate performance of the executed control operations by the virtual vehicle based on the compressed map data for each iteration, and train the machine-learning compression model to compress map data such that the evaluated performance of the executed control operations by the virtual vehicle exceeds a performance threshold.

Abstract: Vehicle control systems and methods for controlling a vehicle using behavior profiles of neighboring vehicles are disclosed. In one embodiment, a vehicle control system of a vehicle includes one or more processors and one or more non-transitory memory modules communicatively coupled to the one or more processors and storing machine-readable instructions that, when executed, cause the one or more processors to perform at least at the following: detect a neighboring vehicle within an environment, determine an autonomous software version executed by the neighboring vehicle, determine a behavior profile associated with the autonomous software version, and control one or more vehicle systems of the vehicle based at least in part on the behavior profile associated with the autonomous software version.

Abstract: Vehicle control systems and methods for controlling a vehicle using behavior profiles of neighboring vehicles are disclosed. In one embodiment, a vehicle control system of a vehicle includes one or more processors and one or more non-transitory memory modules communicatively coupled to the one or more processors and storing machine-readable instructions that, when executed, cause the one or more processors to perform at least at the following: detect a neighboring vehicle within an environment, determine an autonomous software version executed by the neighboring vehicle, determine a behavior profile associated with the autonomous software version, and control one or more vehicle systems of the vehicle based at least in part on the behavior profile associated with the autonomous software version.

Abstract: Methods and systems for determining and presenting virtual safety cages are provided. An example method may involve receiving an instruction for a robotic device to perform a physical action in a physical environment occupied by the robotic device. The method may also involve, responsive to receiving the instruction, and based on one or more parameters of one or more physical components of the robotic device, determining one or more estimated trajectories along which the one or more physical components of the robotic device are estimated to move as the robotic device performs the physical action. The method may further involve, based on the one or more estimated trajectories, determining a virtual representation of a space that the robotic device is estimated to occupy in the physical environment while performing the physical action. The method may then involve providing, into the physical environment, an indication of a location of the space.

Abstract: Methods and computer-program products for evaluating grasp patterns for use by a robot are disclosed. In one embodiment, a method of evaluating grasp patterns includes selecting an individual grasp pattern from a grasp pattern set, establishing a thumb-up vector, and simulating the motion of the manipulator and the end effector according to the selected individual grasp pattern, wherein each individual grasp pattern of the grasp pattern set corresponds to motion for manipulating a target object. The method further includes evaluating a direction of the thumb-up vector during at least a portion of the simulated motion of the manipulator and the end effector, and excluding the selected individual grasp pattern from use by the robot if the direction of the thumb-up vector during the simulated motion is outside of one or more predetermined thresholds. Robots utilizing the methods and computer-program products for evaluating grasp patterns are also disclosed.

Abstract: Methods and computer-program products for evaluating grasp patterns for use by a robot are disclosed. In one embodiment, a method of evaluating grasp patterns includes selecting an individual grasp pattern from a grasp pattern set, establishing a thumb-up vector, and simulating the motion of the manipulator and the end effector according to the selected individual grasp pattern, wherein each individual grasp pattern of the grasp pattern set corresponds to motion for manipulating a target object. The method further includes evaluating a direction of the thumb-up vector during at least a portion of the simulated motion of the manipulator and the end effector, and excluding the selected individual grasp pattern from use by the robot if the direction of the thumb-up vector during the simulated motion is outside of one or more predetermined thresholds. Robots utilizing the methods and computer-program products for evaluating grasp patterns are also disclosed.

Abstract: Methods and computer program products for generating robot grasp patterns are disclosed. In one embodiment, a method for generating robot grasp patterns includes generating a plurality of approach rays associated with a target object. Each approach ray of the plurality of approach rays extends perpendicularly from a surface of the target object. The method further includes generating at least one grasp pattern for each approach ray to generate a grasp pattern set of the target object, calculating a grasp quality score for each individual grasp pattern of the grasp pattern set, and comparing the grasp quality score of each individual grasp pattern with a grasp quality threshold. The method further includes selecting individual grasp patterns of the grasp pattern set having a grasp quality score that is greater than the grasp quality threshold, and providing the selected individual grasp patterns to the robot for on-line manipulation of the target object.

Abstract: Methods and computer-program products for evaluating grasp patterns for use by a robot are disclosed. In one embodiment, a method of evaluating grasp patterns includes selecting an individual grasp pattern from a grasp pattern set, establishing a thumb-up vector, and simulating the motion of the manipulator and the end effector according to the selected individual grasp pattern, wherein each individual grasp pattern of the grasp pattern set corresponds to motion for manipulating a target object. The method further includes evaluating a direction of the thumb-up vector during at least a portion of the simulated motion of the manipulator and the end effector, and excluding the selected individual grasp pattern from use by the robot if the direction of the thumb-up vector during the simulated motion is outside of one or more predetermined thresholds. Robots utilizing the methods and computer-program products for evaluating grasp patterns are also disclosed.

Abstract: Robots, computer program products, and methods for trajectory plan optimization are disclosed. In one embodiment, a method of controlling a robot having a first manipulator and a second manipulator includes receiving a trajectory plan including a plurality of sequential motion segments. The method further includes determining a moveable motion segment, and shifting the moveable motion segment and motion segments subsequent to the moveable motion segment backward in time to a shifted time such that one or more unshifted segments of the trajectory plan occur at a same time as one or more shifted segment segments. The method may further include controlling the robot according to the optimized trajectory plan such that one or more components of the first manipulator are moved concurrently with one or more components of the second manipulator.

Abstract: Methods and computer-program products for evaluating grasp patterns for use by a robot are disclosed. In one embodiment, a method of evaluating grasp patterns includes selecting an individual grasp pattern from a grasp pattern set, establishing a thumb-up vector, and simulating the motion of the manipulator and the end effector according to the selected individual grasp pattern, wherein each individual grasp pattern of the grasp pattern set corresponds to motion for manipulating a target object. The method further includes evaluating a direction of the thumb-up vector during at least a portion of the simulated motion of the manipulator and the end effector, and excluding the selected individual grasp pattern from use by the robot if the direction of the thumb-up vector during the simulated motion is outside of one or more predetermined thresholds. Robots utilizing the methods and computer-program products for evaluating grasp patterns are also disclosed.

Abstract: Robots, computer program products, and methods for trajectory plan optimization are disclosed. In one embodiment, a method of controlling a robot having a first manipulator and a second manipulator includes receiving a trajectory plan including a plurality of sequential motion segments. The method further includes determining a moveable motion segment, and shifting the moveable motion segment and motion segments subsequent to the moveable motion segment backward in time to a shifted time such that one or more unshifted segments of the trajectory plan occur at a same time as one or more shifted segment segments. The method may further include controlling the robot according to the optimized trajectory plan such that one or more components of the first manipulator are moved concurrently with one or more components of the second manipulator.

Abstract: Methods and computer program products for generating robot grasp patterns are disclosed. In one embodiment, a method for generating robot grasp patterns includes generating a plurality of approach rays associated with a target object. Each approach ray of the plurality of approach rays extends perpendicularly from a surface of the target object. The method further includes generating at least one grasp pattern for each approach ray to generate a grasp pattern set of the target object, calculating a grasp quality score for each individual grasp pattern of the grasp pattern set, and comparing the grasp quality score of each individual grasp pattern with a grasp quality threshold. The method further includes selecting individual grasp patterns of the grasp pattern set having a grasp quality score that is greater than the grasp quality threshold, and providing the selected individual grasp patterns to the robot for on-line manipulation of the target object.