Human skill based path generation

A method for human skill based robot path generation, said method comprising:

• providing a training database including a plurality of human-generated robot motion programs, including data defining obstacles which are avoided during robotic execution of each motion program;
• training an encoder/decoder neural network system, including an encoder neural network and a corresponding decoder neural network running on a computing device, using the training database, where the training causes the encoder neural network to learn a distribution of probabilities associated with states and actions and the decoder neural network determines a probability function of actions corresponding with states and probabilities;
• providing obstacle feature data and start and goal positions for a new path to the encoder/decoder neural network system, including providing a point cloud or solid model representation of obstacles in a workspace of the new path to a three-dimensional convolutional neural network (3D CNN), extracting feature vectors characterizing the obstacles using the 3D CNN, and providing the feature vectors to the encoder/decoder neural network system;
• computing a waypoint distribution for the new path by the encoder/decoder neural network system, where the waypoint distribution is a set of candidate waypoints for the new path;
• generating the new path based on the waypoint distribution, where the new path defines motions for a robot to move a tool center point from the start position to the goal position, where generating the new path includes using either a rapidly-exploring random tree (RRT) computation or an optimization-based computation and includes performing a robot-obstacle collision avoidance calculation for each path point;
• sending motion commands to the robot, by a controller, which cause the robot to move the tool center point along the new path;
• and
• executing the motion commands by the robot.

A system for human skill based robot path generation, said system comprising:

• a robot in communication with a robot controller;
• and
• a computing device having a processor and memory, said computing device being configured with one or more algorithm to perform steps including;
• accessing a training database including a plurality of human-generated robot motion programs, including data defining obstacles which are avoided during robotic execution of each motion program;
• training an encoder/decoder neural network system, including an encoder neural network and a corresponding decoder neural network, using the training database;
• providing obstacle feature data and start and goal positions for a new path to the encoder/decoder neural network system;
• computing a waypoint distribution for the new path by the encoder/decoder neural network system, where the waypoint distribution is a set of candidate waypoints for the new path;
• and
• generating the new path based on the waypoint distribution, where the new path defines motions for the robot to move a tool center point from the start position to the goal position, including performing a robot-obstacle collision avoidance calculation for each path point,
• where the new path is provided to the controller, and the robot executes the new path based on motion commands from the controller.