Legged high-dexterity self-balancing capable robot actor

A method of generating control animations, motion parameters, or scripts for a self-balancing, legged robot, comprising:

• animating a performance for a legged robot with a predefined hardware configuration;
• verifying whether the animated performance complies with physical constraints, wherein when the animated performance is noncompliant with the physical constraints the animating is repeated;
• and
• running a simulation of the animated performance based on the predefined hardware configuration to determine whether the legged robot balances, wherein the animating is repeated when the legged robot loses balance during the running of the simulation.

A robot for bringing characters to life in the physical world, comprising:

• a pair of movable legs;
• memory storing authored animation content generated to correspond with a screen-based character;
• sensors perceiving inputs from a physical environment in which the robot is positioned;
• and
• a controller including a behavior engine processing the authored animation content along with the inputs from the sensors to generate motion commands to control motion of the pair of movable legs,
• wherein the controller performs motion synthesis prior to generating the motion commands,
• wherein the behavior engine selects and places in sequential order a set of movements in the authored animation content based on the inputs from the sensors,
• wherein a set of actuators in the pair of legs operate in response to the motion commands,
• wherein the legs each comprise an upper leg housing and a lower leg housing coupled via a knee joint,
• wherein the upper leg housing includes sidewalls defining a hollow interior space,
• wherein channels are provided in the upper leg housing for receiving two or more of the actuators,
• wherein the sidewalls include air inlets for drawing air into the hollow interior space and air outlets for exhausting air out of the hollow interior space, and
• wherein a stack of spaced apart metal fins are positioned within the hollow interior space that are coupled to sidewalls defining the channels and that define air plenums for air between the air inlets and the air outlets.

A robot for bringing characters to life in the physical world, comprising:

• movable components;
• memory storing a set of animations, motion parameters, or scripts defining a set of gestures and a set of movements for each of a plurality of moods or emotions, wherein the set of animations or scripts are generated based on one of the characters;
• and
• a controller located onboard the robot,
• wherein the robot is untethered and free-ranging in a physical space,
• wherein the controller determines, in real time, a present mood or emotion associated with the robot,
• wherein the controller selects one of the sets of gestures or one of the sets of movements based on the present mood or emotion,
• wherein the controller operates a set of actuators to operate the movable components using the selected one of the sets of gestures or the sets of movements, whereby the robot is operated to provide an illusion of life,
• wherein the movable components comprise two or more legs each comprising an upper leg housing and a lower leg housing coupled via a knee joint,
• wherein the upper leg housing includes sidewalls defining a hollow interior space,
• wherein channels are provided in the upper leg housing for receiving two or more of the actuators including an ankle pitch actuator, and
• wherein two belts extend from the upper leg housing and the ankle pitch actuator over a pulley in the lower leg housing with a configuration that transmits torque of the ankle pitch actuator across the knee joint to provide assistive torque during stance using reversed gastrocnemius.

A robot for bringing characters to life in the physical world, comprising:

• movable components;
• memory storing a set of animations, motion parameters, or scripts defining a set of gestures and a set of movements for each of a plurality of moods or emotions, wherein the set of animations or scripts are generated based on one of the characters;
• and
• a controller located onboard the robot,
• wherein the robot is untethered and free-ranging in a physical space,
• wherein the controller determines, in real time, a present mood or emotion associated with the robot,
• wherein the controller selects one of the sets of gestures or one of the sets of movements based on the present mood or emotion,
• wherein the controller operates a set of actuators to operate the movable components using the selected one of the sets of gestures or the sets of movements, whereby the robot is operated to provide an illusion of life, and
• wherein at least a subset of the actuators each includes a clutch acting as a self-resetting mechanical fuse when transient torques exceed actuator component limits to handle impact events.

A robot for bringing characters to life in the physical world, comprising:

• a pair of movable legs;
• memory storing authored animation content generated to correspond with a screen-based character;
• sensors perceiving inputs from a physical environment in which the robot is positioned;
• and
• a controller including a behavior engine processing the authored animation content along with the inputs from the sensors to generate motion commands to control motion of the pair of movable legs,
• wherein the controller performs motion synthesis prior to generating the motion commands,
• wherein the behavior engine selects and places in sequential order a set of movements in the authored animation content based on the inputs from the sensors,
• wherein a set of actuators in the pair of legs operate in response to the motion commands,
• wherein the legs each comprise an upper leg housing and a lower leg housing coupled via a knee joint,
• wherein the upper leg housing includes sidewalls defining a hollow interior space,
• wherein channels are provided in the upper leg housing for receiving two or more of the actuators including an ankle pitch actuator, and
• wherein two belts extend from the upper leg housing and the ankle pitch actuator over a pulley in the lower leg housing with a configuration that transmits torque of the ankle pitch actuator across the knee joint to provide assistive torque during stance using reversed gastrocnemius.

A robot for bringing characters to life in the physical world, comprising:

• a pair of movable legs;
• memory storing authored animation content generated to correspond with a screen-based character;
• sensors perceiving inputs from a physical environment in which the robot is positioned;
• and
• a controller including a behavior engine processing the authored animation content along with the inputs from the sensors to generate motion commands to control motion of the pair of movable legs,
• wherein the controller performs motion synthesis prior to generating the motion commands,
• wherein the behavior engine selects and places in sequential order a set of movements in the authored animation content based on the inputs from the sensors,
• wherein a set of actuators in the pair of legs operate in response to the motion commands, and
• wherein at least a subset of the actuators each includes a clutch acting as a self-resetting mechanical fuse when transient torques exceed actuator component limits to handle impact events.