Apparatus, methods and articles to facilitate motion planning in environments having dynamic obstacles

A motion planning method of operation in a processor-based system to perform motion planning via planning lattices, where each planning lattice respectively comprises a plurality of nodes and edges, each node representative of a state of a primary vehicle which operates in an environment that includes one or more other vehicles and other obstacles, and each edge representative of a transition between a respective pair of the nodes, the method comprising:

• representing, by the processor-based system, the primary vehicle as a respective oriented bounding box;
• for at least one of a plurality of the edges of the planning lattice,
• performing, by the processor-based system, oriented bounding box collision detection to determine whether the primary vehicle represented by the oriented bounding box will collide with representations of other vehicles or other obstacles in transitioning between a pair of states, the state of the pair represented by respective nodes of a pair of nodes that are connected by the respective edge of the planning lattice;
• and
• setting, by the processor-based system, a cost of the respective edge of a planning lattice to reflect at least one of the detected collision or absence thereof;
• and
• wherein performing oriented bounding box collision detection includes performing oriented bounding box collision detection at each of at least one point along a respective trajectory of the primary vehicle using a set of fitted functions to represent the respective trajectory of the primary vehicle corresponding to the respective edge of the planning lattice, and further comprising:
• performing, by the processor-based system, a least cost analysis of the planning lattice;
• and
• implementing, by the processor-based system, a resultant motion plan by the primary vehicle based at least in part on the least cost analysis.

A processor-based system to perform motion planning via planning lattices, where each planning lattice respectively comprises a plurality of nodes and edges, each node representative of a state of a primary vehicle which operates in an environment that includes one or more other vehicles and other obstacles, and each edge representative of a transition between a respective pair of the nodes, the processor-based system comprising:

• at least one processor;
• at least one nontransitory processor-readable medium communicatively coupled to the at least one processor and that stores processor-executable instructions that, when executed by the at least one processor, cause the at least one processor to:
• represent the primary vehicle as a respective oriented bounding box;
• represent at least one other obstacle as a representation of a curve in the form of a spline, a b-spline, or a polynomial, and perform the oriented bounding box collision detection as oriented bounding box-curve collision detection to determine whether the respective oriented bounding box that represents the primary vehicle will collide with the curve that represents the at least one of the other obstacles in transitioning between the pair of states represented by respective nodes that are connected by the respective edge of the planning lattice;
• and
• for at least one of a plurality of the edges of the planning lattice,
• perform oriented bounding box collision detection to determine whether the primary vehicle represented by the oriented bounding box will collide with representations of other vehicles or other obstacles in transitioning between a pair of states, the state of the pair represented by respective nodes of a pair of nodes that are connected by the respective edge of the planning lattice;
• and
• set a cost of the respective edge of a planning lattice to reflect at least one of the detected collision or absence thereof;
• perform a least cost analysis of the planning lattice;
• and
• implement a resultant motion plan by the primary vehicle based at least in part on the least cost analysis.

A processor-based system to perform motion planning via planning lattices, where each planning lattice respectively comprises a plurality of nodes and edges, each node representative of a state of a primary vehicle which operates in an environment that includes one or more other vehicles and other obstacles, and each edge representative of a transition between a respective pair of the nodes, the processor-based system comprising:

• at least one processor;
• at least one nontransitory processor-readable medium communicatively coupled to the at least one processor and that stores processor-executable instructions that, when executed by the at least one processor, cause the at least one processor to:
• represent the primary vehicle as a respective oriented bounding box;
• for at least one of a plurality of the edges of the planning lattice,
• perform oriented bounding box collision detection to determine whether the primary vehicle represented by the oriented bounding box will collide with representations of other vehicles or other obstacles in transitioning between a pair of states, the state of the pair represented by respective nodes of a pair of nodes that are connected by the respective edge of the planning lattice;
• and
• set a cost of the respective edge of a planning lattice to reflect at least one of the detected collision or absence thereof,
• wherein the processor-executable instructions, when executed by the at least one processor, cause the at least one processor to perform the oriented bounding box collision detection at each of at least one point along a respective trajectory of the primary vehicle via a set of fitted functions to represent the respective trajectory of the primary vehicle corresponding to the respective edge of the planning lattice;
• wherein the processor-executable instructions, when executed by the at least one processor, cause the at least one processor further to:
• perform a least cost analysis of the planning lattice;
• and
• implement a resultant motion plan by the primary vehicle based at least in part on the least cost analysis.

A motion planning method of operation in a processor-based system to perform motion planning via planning lattices, where each planning lattice respectively comprises a plurality of nodes and edges, each node which characterizes a state of a primary vehicle, the primary vehicle which operates in an environment that includes one or more other vehicles and other obstacles, and each edge represents a trajectory of the primary vehicle between states represented by a respective pair of the nodes, the method comprising:

• during a runtime, for each of a number of the edges of a planning lattice,
• representing, by the processor-based system, the trajectory corresponding to the respective edge as a set of fitted functions,
• performing, by the processor-based system, collision detection using the set of fitted functions to determine whether a representation of the primary vehicle collides with one or more representations of other vehicles or other obstacles;
• and
• setting, by the processor-based system, a cost of the respective edge of the planning lattice to reflect at least one of the determined collision or absence thereof in transitioning between a pair of states represented by respective nodes that are connected by the respective edge of the planning lattice.