Method and apparatus for improved sampling-based graph generation for online path planning by a robot

A method of operation by an online path planner in controlling motion of a robot within an environment, the method comprising:

• performing a sampling procedure that selects partial poses of the robot within the environment, such that the sampling procedure works at a dimensionality lower than theDegrees of Freedom (DoF) of the robot, each partial robot pose specifying values for fewer than all configuration parameters needed to define a complete pose of the robot within the environment;
• and
• converting each partial robot pose into a corresponding complete robot pose by, for each partial robot pose, determining values for configuration parameters not accounted for in the partial robot pose according to a distance function that relates the partial robot pose to a set of two or more reference robot poses;
• and
• taking the complete robot poses as nodes of a graph stored as a data structure held in memory, and determining edges between respective pairs of the nodes, each edge of the graph being a feasible path between a particular pair of the nodes;
• and
• commanding movement of the robot in real time, to follow a path within the environment determined according to the graph.

A computer system for online path planning for a robot based on a graph comprising a data structure comprising a collection of nodes and edges, with each node being a respective robot pose in a sampling frame of reference and each edge being a feasible path between a respective pair of the nodes, the computer system comprising:

• interface circuitry configured to receive data that identifies a set of two or more reference robot poses associated with task performance by the robot;
• and
• processing circuitry configured to:
• generate each node as a corresponding configuration vector comprising a plurality of vector elements, with each vector element being a value for a corresponding one among a plurality of configuration parameters representing the degrees of freedom (DoF) of the robot, based on:
• setting a value for each of one or more vector elements in a first subset of the vector elements independent of the set of two or more reference robot poses, the first subset defining a partial robot pose;
• setting a value for each of one or more vector elements in a disjoint second subset of the vector elements according to a distance function that relates the partial robot pose to one or more of the reference robot poses;
• and
• determining the edges according to path feasibility between respective pairs of the nodes;
• save the graph to a computer-readable medium;
• and
• perform online path planning based on the saved graph and correspondingly generate real-time commands causing the robot to move along a path determined via the online path planning.