Robotic training apparatus and methods

Abstract

Apparatus and methods for training of robotic devices. Robotic devices may be trained by a user guiding the robot along target trajectory using an input signal. A robotic device may comprise an adaptive controller configured to generate control commands based on one or more of the user guidance, sensory input, and/or performance measure. Training may comprise a plurality of trials. During first trial, the user input may be sufficient to cause the robot to complete the trajectory. During subsequent trials, the user and the robot's controller may collaborate so that user input may be reduced while the robot control may be increased. Individual contributions from the user and the robot controller during training may be may be inadequate (when used exclusively) to complete the task. Upon learning, user's knowledge may be transferred to the robot's controller to enable task execution in absence of subsequent inputs from the user

What is claimed: 
     
         1 .- 23 . (canceled) 
     
     
         24 . A robotic apparatus, comprising:
 a platform comprising one or more controllable elements; and   a controller communicatively coupled to the platform to operate individual ones of the one or more controllable elements, the controller configured to:   receive a first input from a human;   analyze the first input and cause the platform to execute a first action in accordance with the first input, wherein the analysis of the first input comprises a determination of a deviation between the first action and a target action; and   receive a second input from a human subsequent to the receipt of the first input, the second input being configured to cause a corrected action by the platform, the corrected action being characterized at least in part by a lower deviation from the target action.   
     
     
         25 . The robotic apparatus of  claim 24 , wherein the first action is based on the operation of the individual ones of the one or more controllable elements of the platform. 
     
     
         26 . The robotic apparatus of  claim 24 , wherein the first input and the second input are sensory signals configured to convey information associated with one or both of an environment of the robotic apparatus and a state of the platform. 
     
     
         27 . The robotic apparatus of  claim 24 , wherein the controller is further configured to generate a predicted control output based at least in part on the first input, wherein the first action is determined based at least in part on the predicted control output. 
     
     
         28 . The robotic apparatus of  claim 24 , wherein the controller is further configured to search a predetermined table correlating a plurality of human inputs, a plurality of sensory signal characteristics, and a plurality of predicted control outputs, wherein the first action is determined based at least in part on at least one of the predicted control outputs correlated to the first input. 
     
     
         29 . The robotic apparatus of  claim 24 , wherein the platform further comprises an actuator configured to actuate the one or more controllable elements. 
     
     
         30 . The robotic apparatus of  claim 24 , wherein the target action is based at least in part on a training input by a human. 
     
     
         31 . The robotic apparatus of  claim 24 , further comprising a user interface configured to receive inputs from a human. 
     
     
         32 . A method of operating a robot, comprising:
 receiving a first input from a human;   analyzing the first input and causing one or more controllable elements to execute a first action in accordance with the first input, wherein analyzing the first input comprises determining a deviation between the first action and a target action; and   receiving a second input from a human subsequent to the receipt of the first input, the second input being configured to cause a corrected action by the platform, the corrected action being characterized at least in part by a lower deviation from the target action.   
     
     
         33 . The method of  claim 32 , further comprising operating individual ones of the one or more controllable elements of the platform based at least in part on the first action. 
     
     
         34 . The method of  claim 32 , further comprising generating a sensory signal based at least in part upon determined information of one or both of an environment of the robot and a state of the one or more controllable elements. 
     
     
         35 . The method of  claim 34 , further comprising determining one or more predicted control outputs based at least in part on the sensor signal. 
     
     
         36 . The method of  claim 32 , further comprising generating a predicted control output based at least in part on the first input, and determining the first action based at least in part on the predicted control output. 
     
     
         37 . The method of  claim 36 , further comprising generating a combined output by combining the predicted control output and the first input. 
     
     
         38 . The method of  claim 37 , further comprising performing a supervised learning process based at least in part on the combined output. 
     
     
         39 . The method of  claim 32 , further comprising effectuating the corrected action based at least in part on a cooperative interaction with a human, wherein the cooperative interaction characterized by a plurality of iterations. 
     
     
         40 . A computerized robotic controller apparatus, comprising:
 one or more processors configured to execute computer program modules to cause the one or more processors to:   for a given iteration of a plurality of iterations:   generate a first signal based on a sensory context;   receive a second signal from a user associated with a target action configured based on the sensory context; and   utilize the first signal based on the sensory context and the second signal from the user to perform an action;   wherein:   the plurality of iterations is configured to occur within a time interval characterized by a predetermined duration;   a subsequent iteration of the plurality of iterations is configured to cause the action to deviate less from the target action; and   upon expiration of the predetermined duration, the first signal is configured to perform the action absent the second signal.   
     
     
         41 . The apparatus of  claim 40 , wherein:
 individual ones of the plurality of iterations are spaced by time intervals that are at ten percent of the predetermined duration or less; and   the plurality of iterations include more than seven iterations.   
     
     
         42 . The apparatus of  claim 40 , individual ones of the plurality of iterations further include determining a performance measure based at least in part on the action and the target action. 
     
     
         43 . The apparatus of  claim 42 , wherein the one or more processors are further configured to adjust a learning parameter based at least in part on the performance measure of the individual ones of the plurality of iterations, and the action is further based on the learning parameter.