Robot plan online adjustment

A system comprising one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:

• receiving an initial plan for performing a particular task with a robot, wherein the initial plan is computed offline and defines a plurality of initial waypoints defining an initial path from a starting point to an end point, and wherein each initial waypoint is associated with parameters including a target position, a target velocity, and a target acceleration;
• providing a first command to the robot to drive the robot to follow the initial path;
• after providing the first command to the robot to drive the robot to follow the initial path, generating, during online operation of the robot, an alternative path from the initial path, including:
• generating a plurality of alternative paths from the starting point to the end point, including performing respective modifications to at least one of the parameters of one or more initial waypoints in the initial plan to generate alternative waypoints, each alternative path of the plurality of alternative paths comprising a plurality of alternative waypoints,
• evaluating each alternative path of the plurality of alternative paths according to a simulated total time duration required for the robot to traverse the alternative path from the starting point to the end point, and
• selecting an alternative path having a total time duration that is less than a total time duration of the initial plan;
• and
• providing, during online operation of the robot, a second command to the robot to drive the robot to follow one or more alternative waypoints of the selected alternative path instead of the initial path.

A non-transitory computer storage medium encoded with instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:

• receiving an initial plan for performing a particular task with a robot, wherein the initial plan is computed offline and defines a plurality of initial waypoints defining an initial path from a starting point to an end point, and wherein each initial waypoint is associated with parameters including a target position, a target velocity, and a target acceleration;
• providing a first command to the robot to drive the robot to follow the initial path;
• after providing the first command to the robot to drive the robot to follow the initial path, generating, during online operation of the robot, an alternative path from the initial path, including:
• generating a plurality of alternative paths from the starting point to the end point, including performing respective modifications to at least one of the parameters of one or more initial waypoints in the initial plan to generate alternative waypoints, each alternative path of the plurality of alternative paths comprising a plurality of alternative waypoints,
• evaluating each alternative path of the plurality of alternative paths according to a simulated total time duration required for the robot to traverse the alternative path from the starting point to the end point, and
• selecting an alternative path having a total time duration that is less than a total time duration of the initial plan;
• and
• providing, during online operation of the robot a second command to the robot to drive the robot to follow one or more alternative waypoints of the selected alternative path instead of the initial path.