Abstract: An information processing apparatus includes a crowd characteristic estimator and a plan controller. The crowd characteristic estimator estimates information regarding a characteristic of a crowd on the basis of a sensing result of an external environment. The crowd is a collection of individuals present in the external environment. The plan controller controls a creation mode of an action plan for a movable body in the external environment on the basis of at least the information regarding the characteristic.

Abstract: A mobile body control device includes: an observation trajectory acquisition unit that acquires, a trajectory along which the subject walks; a basic trajectory acquisition unit that acquires, a basic trajectory that matches the observation trajectory; a subject predicted trajectory acquisition unit that acquires a trajectory along which the subject walks after the current time point based on the specific basic trajectory; a prediction reliability calculation unit that calculates a prediction reliability based on uncertainty added along the subject predicted trajectory; a target trajectory acquisition unit that acquires, a trajectory formed by a target position of a mobile body set ahead of the subject in a traveling direction of the subject; and a control-command-sequence-generating unit that generates a control command sequence using an evaluation function that uses as input the subject predicted trajectory, prediction reliability, target trajectory, and a control trajectory.

Abstract: A bifurcated navigation control system defines a path whereby a manipulator cart is to navigate from an initial location to a target location, and identifies a navigation condition associated with a navigation of the manipulator cart along the path. Based on the navigation condition, the bifurcated navigation control system defines a propulsion limitation for the manipulator cart during the navigation of the manipulator cart along the path. The bifurcated navigation control system directs the manipulator cart to navigate along at least part of the path in a bifurcated navigation control mode in which the bifurcated navigation control system is configured to autonomously control a steering of the manipulator cart while allowing operator control of a propulsion of the manipulator cart in accordance with the propulsion limitation. Corresponding methods and systems are also disclosed.

Abstract: A computer implemented method for determining a configuration of a medical robotic arm, wherein the configuration comprises a pose of the robotic arm and a position of a base of the robotic arm, comprising the steps of: acquiring treatment information data representing information about the treatment to be performed by use of the robotic arm; acquiring patient position data representing the position of a patient to be treated; and calculating the configuration from the treatment information data and the patient position data.

Abstract: An apparatus for automated collision avoidance includes a sensor configured to detect an object of interest, predicting a representation of the object of interest at a future point in time, calculating an indication of a possibility of a collision with the object of interest based on the representation of the object of interest at the future point in time, and executing a collision avoidance action based on the indication.

Abstract: A robot system is provided and includes a robot body that performs predetermined work together with a user, and a control device that controls the robot body. The control device selects, in accordance with special information of the user, an action to be performed by the robot body during the predetermined work.

Abstract: A robot monitoring apparatus includes a position monitoring unit having a position monitoring function of calculating an attitude of a robot, and monitoring whether or not the robot interferes with a virtual safety fence, and a speed monitoring unit having a speed monitoring function of calculating and monitoring a speed of the robot. The apparatus has a first monitor mode in which, when the robot moves in an automatic operation mode, the robot is stopped when the robot interferes with the virtual safety fence in the position monitoring function and when the speed of the robot exceeds a first upper limit value in the speed monitoring function, and a second monitor mode in which, when the robot moves in a manual operation mode, the position monitoring function is disabled and the robot is stopped when the speed of the robot exceeds a second upper limit value smaller than the first upper limit value in the speed monitoring function.

Abstract: A provider institution computing system determining a goal of the customer of the provider institution, determining context information related to the goal, generating a first set of sub-goals for the goal of the customer, the first set of sub-goals defining a first group of steps required to be met by the customer to reach the goal, determining a current status of the customer relating to the first set of sub-goals and the goal, generating a second set of sub-goals for the goal of the customer corresponding to a first change in the context information, the second set of sub-goals being different than the first set of sub-goals, wherein the goal remains unchanged upon generating the second set of sub-goals, and responsive to determining that at least one transaction fails to conform with the goal, blocking, by the artificial intelligence circuit, the at least one transaction.

Abstract: A technique for automatically finding a collision-free return-to-home path for a robot. The technique includes running a simulated virtual 3D environment which emulates the physical robot and workcell in real time, including the positions and poses of all robots, workpieces and obstacles in the workcell. Upon request by an operator, a return-to-home path search is executed based on the virtual 3D environment, where the path search calculates a solution which moves the robot from a current position to its home or recovery position while avoiding collisions with other robots, workpieces or objects in the workcell. In addition to collision avoidance, the path search considers other constraints such as prohibited zones in the workspace and robot joint positions. When the recovery path is computed, the solution program is sent back to the physical environment for execution by the physical robot.

Abstract: A method and system for motion planning for robots with a redundant degree of freedom. The technique computes a collision avoidance motion plan for a robot with a redundant degree of freedom, without artificially constraining the extra degree of freedom. The motion planning is formulated as a quadratic programming optimization calculation having a multi-component objective function and a collision avoidance constraint function. The formulation is efficient enough to compute the motion plan in real time at every robot control cycle. The collision avoidance constraint ensures clearance of all parts of the robot from both static and dynamic obstacles. Objective function terms include minimizing path deviation, joint velocity regularization and robot configuration or pose regularization. Weighting factors on the terms of the objective function are changeable for each control cycle calculation based on obstacle proximity conditions at the time.

Abstract: A system includes a computer on board a vehicle and a server remote from the vehicle. The computer and the server each store a network graph of actions performable by components of the vehicle that includes costs of transitioning between the actions. The computer is programmed to log a sequence of the actions; upon determining that a sub-sequence of the sequence constitutes a lowest-cost path from the beginning action to the final action based on the network graph, store a reduced sequence including a beginning action and a final action of the sub-sequence ordered immediately after the beginning action, the reduced sequence excluding at least one intermediate action from the sub-sequence; and transmit the reduced sequence to a server remote from the vehicle. The server is programmed to determine the sequence from the reduced sequence based on the network graph.

Abstract: A robot interference checking motion planning technique using swept volume deformation. A rapidly-exploring random tree (RRT) algorithm generates random sample nodes between a start point and a goal point. Each sample node is evaluated by checking for robot-obstacle interference along a path segment to the node. If an interference exists along the path segment, a swept volume of the segment is used to identify a critical posture where the interference is greatest, and obstacle interference points are used to define a virtual force applied to the robot links to modify the path segment to alleviate the interference condition. A swept volume of the modified path segment is computed and evaluated. If the modified swept volume is collision-free and the modified path segment motion plan meets robot joint range criteria, the modified path segment and the sample node are added to the overall robot motion program.

Abstract: Systems and methods may support generation and use of simulation signature keys for robotic simulations. A computer-aided manufacturing (CAM) system may access robotic path data for a robot that includes target locations and corresponding motion parameters for the robot. The CAM system may partition the robotic path data into simulation events according to event partitioning criteria such that each simulation event includes one or multiple target locations and corresponding motion parameters and further generate a respective signature simulation key for each partitioned simulation event. The simulation signature key may provide a lookup key into a motion planning data cache. When an entry exists in the motion planning data cache for a simulation signature key generated for a simulation event, the CAM system may retrieve cached motion planning data for the simulation event. When an entry does not exist, the CAM system may obtain motion planning data through robotic simulation.

Abstract: The invention relates to a method for controlling a robotic device (50) with modified control commands transmitted over a wireless network, wherein the robotic device (50) comprises a plurality of joints (53), wherein each joint represents one degree of freedom of the robotic device, the method comprising at a trajectory modification entity (100): —determining a load of the wireless network (30), —receiving a plurality of control commands controlling a planned trajectory of the robotic device (50) from a robotic control entity (70), each of the control commands configured to control one degree of freedom of a first number of degrees of freedom addressed by the plurality of control commands, —determining a reduced number of degrees of freedom for the modified control commands smaller than the first number based on the determined load, —determining the modified control commands based on the reduced number of degrees of freedom, wherein the modified control commands address a limited number of degrees of freedom

Abstract: An information processor calculates, for a robot hand including a plurality of fingers, a gripping pose at which the robot hand grips a target object. The information processor includes a candidate single-finger placement position detector that detects, based on three-dimensional measurement data obtained through three-dimensional measurement of the target object and hand shape data about a shape of the robot hand, candidate placement positions for each of the plurality of fingers of the robot hand, a multi-finger combination searcher that searches for, among the candidate placement positions for each of the plurality of fingers, a combination of candidate placement positions to allow gripping of the target object, and a gripping pose calculator that calculates, based on the combination of candidate placement positions for each of the plurality of fingers, a gripping pose at which the robot hand grips the target object.

Abstract: A robot control system determines which of a number of discretizations to use to generate discretized representations of robot swept volumes and to generate discretized representations of the environment in which the robot will operate. Obstacle voxels (or boxes) representing the environment and obstacles therein are streamed into the processor and stored in on-chip environment memory. At runtime, the robot control system may dynamically switch between multiple motion planning graphs stored in off-chip or on-chip memory. The dynamically switching between multiple motion planning graphs at runtime enables the robot to perform motion planning at a relatively low cost as characteristics of the robot itself change.

Abstract: An arithmetic device is an arithmetic device configured to output information to a moving object that automatically moves. The arithmetic device includes: a target object information acquisition unit configured to acquire, from a sensor provided in a place other than to the moving object, a detection result of position information of a target object, the position information being information related to position and orientation of the target object; a route setting unit configured to set, based on the position information of the target object, a route to a target position at which predetermined position and orientation relative to the target object are reached; and an information output unit configured to output information of the route to the moving object.

Abstract: Described in detail herein is an automated fulfillment system including a computing system programmed to receive requests from disparate sources for physical objects disposed at one or more locations in a facility. The computing system can combine the requests, and group the physical objects in the requests based on object types or expected object locations. Autonomous robot devices can receive instructions from the computing system to retrieve a group of the physical objects and deposit the physical objects in storage containers.

Abstract: A method, computer program and equipment for controlling a crane and method for updating a crane. A crane is controlled or the crane is updated to be controlled by: a) forming video image at a lifting point of a crane; b) determining a current and a future position of a lifting member or of a load with respect to the formed video image; c) combining to the formed video image an indication of the future position of the lifting member or of the load; and d) performing repeatedly again steps b) and c) for real time indicating of the future position of the lifting member or of the load.

Abstract: A computer-automated separation rules compliance method is disclosed. Separation rules that establish separation distance requirements between objects in a three-dimensional (3D) virtual environment are defined. Sample locations associated with at least some of the objects in the 3D virtual environment are specified. One or more of proximity and/or collision analysis is performed on the sample locations to determine separation distances between the objects. The determined separation distances are compared to the separation distance requirements. Objects in the 3D virtual environment that violate the separation rules based on said comparing are identified.

Abstract: A system for detecting an accident risk in a working place that can determine a work stage of the working place where various kinds of works, such as a cargo loading and unloading workshop, are performed or determine whether or not there is a risk that an accident may occurs in the working place is disclosed. In the disclosed system for detecting the accident risk in the working place, a camera is used to photograph the working place, only information about an object recognized by analyzing the image is transmitted to the server without transmitting the photographed image to the server, thereby minimizing an amount of information transmitted from the camera to the server.

Abstract: Any technical error with robotic arms that are used to automatically perform object packing can affect quality and efficiency with which the packing is being carried out, and this in turn affects space utilization when a large quantity of objects are to be accommodated in tight packing spaces. This disclosure relates generally to automated object packing and more specifically to an object packing mechanism in which corrections are made when placement of object is identified as violating one or more regulations. The system packs objects by calculating ICP-BCP pairs for each empty space in a packing space. After packing each object, the system checks whether placement of the object violates one or more regulations, and if any violation is found, then the system determines and executes one or more corrective action to correct placement of the object that violates the regulation.

Abstract: A robot control device controls a robot. The robot control device includes an image recognition processing unit, a robot control processing unit, and a monitoring processing unit. The image recognition processing unit recognizes first information on the basis of measurement data in a monitoring area obtained from a vision sensor, the first information being information about a human present in the monitoring area. The robot control processing unit controls the motion of the robot in accordance with a motion program for moving the robot. On the basis of surrounding object data and the first information obtained from the image recognition processing unit, the monitoring processing unit determines a possibility of pinching of the human between the robot and a surrounding object. The surrounding object data is data indicating three-dimensional disposition states of the robot and a surrounding object that is an object other than the robot in the monitoring area.

Abstract: An information processing apparatus for a travel network including travel paths and reference regions coupling the paths includes a first scheduler configured to generate a first schedule including (1) a sequence of representative regions allowing a first mobile object to pass through the representative regions, and (2) timings of passage through the representative regions, based on move command information for the first mobile object to travel to a first reference region; and a second scheduler selecting a first representative region from among the representative regions included in the first schedule according to a travel situation of the first mobile object, and generate a second schedule including (1) a first route allowing the first mobile object to travel in a first area including the first representative region, and (2) timings for the first mobile object to pass through the reference regions included in the first route.

Abstract: A coordination system for a handling device including a plurality of kinematic chains is provided. Each of the kinematic chains being movable in a workspace. At least two of the workspaces having an overlap, the kinematic chains being designed to carry out a work movement based on a work command. The coordination system includes a trajectory planning module and a control module. The control module is designed to activate the kinematic chains to carry out the work movement based on trajectory data. The trajectory planning module is designed to determine the trajectory data to carry out the work movement and to provide the trajectory data to the control module. The trajectory planning module is designed, if a further work command is provided while the work movement is being carried out, to replan the trajectory data into replanned trajectory data and provide these data to the control module.

Abstract: A path determination device and so forth are provided which can determine a path of a robot such that an autonomous mobile robot smoothly moves to a destination while avoiding an interference with a traffic participant even under a traffic environment such as a crowd. A path determination device (1) determines a temporary moving velocity command v_cnn, by using a CNN, such that an interference between a robot (2) and a traffic participant is avoided and determines a moving velocity command v of the robot (2), by using a DWA, such that in a case where the robot (2) is assumed to move from a present position by the temporary moving velocity command v_cnn, an objective function G(v) including a distance dist to the traffic participant closest to the robot (2) and the temporary moving velocity command v_cnn of the robot (2) as independent variables becomes a maximum value.

Abstract: A predictive system and process that predicts safety system activation in industrial environments when collaborative robots (COBOTs), automated guidance vehicles (AGVs), and other robots (individual or collectively “robots”) are interacting (i) between one another or (ii) between a robot and human. As provided herein, the predictive system is not meant to substitute traditional safety systems, but rather to detect and classify robot-to-robot and robot-to-human interactions and potential interactions thereof so as to limit or avoid those interactions altogether, thereby increasing safety and efficiency of the robots.

Abstract: Techniques for naming devices via voice commands are described herein. For instance, a user may issue a voice command to a voice-controlled device stating, “you are the kitchen device”. Thereafter, the device may respond to voice commands directed, by name, to this device. For instance, the user may issue a voice command requesting to “play music on my kitchen device”. Given that the user has configured the device to respond to this name, the device may respond to the command by outputting the requested music.

Abstract: The present invention relates to a near-site robotic construction system. The system includes a work station situated on a near-site position in a close proximity to a building foundation on which a building is under construction and providing shelter and workspace for at least one robot to work; and a computer-assisted cloud based near-site robotic construction platform installed on a cloud server system and configured to provide for a user to operate through a web browser, import and extract a building information modelling data, and plan a predetermined motion command set partly based on the extracted building information modelling data, wherein the at least one robot is configured to work in accordance with the predetermined motion command set to prefabricate a plurality of components for the building in the work station on the near-site position.

Abstract: A distance-measuring system includes: a first ranging sensor; a second ranging sensor that covers a detection range more limited than a detection range covered by the first ranging sensor and has resolution higher than resolution of the first ranging sensor; and an image processing device. The image processing device includes: a machine detector that detects a position of a mobile machine, based on a first sensed result obtained by the first ranging sensor; a sensor controller that controls a detection position to be detected by the second ranging sensor to detect the position of the mobile machine which has been detected; and a safety determiner that controls the mobile machine, based on a second sensed result obtained by the second ranging sensor.

Abstract: An interactive autonomous robot is configured for deployment within a social environment. The disclosed robot includes a show subsystem configured to select between different in-character behaviors depending on robot status, thereby allowing the robot to appear in-character despite technical failures. The disclosed robot further includes a safety subsystem configured to intervene with in-character behavior when necessary to enforce safety protocols. The disclosed robot is also configured with a social subsystem that interprets social behaviors of humans and then initiates specific behavior sequences in response.

Abstract: A controller controls a motion of an object performing a task for changing a state of the object from a start state to an end state while avoiding collision of the object with an obstacle according to an optimal trajectory determined by solving an optimization problem of the dynamics of the object producing an optimal trajectory for performing the task subject to constraints on a solution of first-order stationary conditions modeling a minimum distance between the convex hull of the object and the convex hull of the obstacle using complementarity constraints.

Abstract: A robot interference checking motion planning technique using point sets. The technique uses CAD models of robot arms and obstacles and converts the CAD models to 3D point sets. The 3D point set coordinates are updated at each time step based on robot and obstacle motion. The 3D points are then converted to 3D grid space indices indicating space occupied by any point on any part. The 3D grid space indices are converted to 1D indices and the 1D indices are stored as sets per object and per time step. Interference checking is performed by computing an intersection of the 1D index sets for a given time step. Swept volumes are created by computing a union of the 1D index sets across multiple time steps. The 1D indices are converted back to 3D coordinates to define the 3D shapes of the swept volumes and the 3D locations of any interferences.

Abstract: Control systems for industrial machinery (e.g., robots) or other devices such as medical devices utilize a safety processor (SP) designed for integration into safety applications and computational components that are not necessarily safety-rated. The SP monitors performance of the non-safety computational components, including latency checks and verification of identical outputs. One or more sensors send data to the non-safety computational components for sophisticated processing and analysis that the SP cannot not perform, but the results of this processing are sent to the SP, which then generates safety-rated signals to the machinery or device being controlled by the SP. As a result, the system may qualify for a safety rating despite the ability to perform complex operations beyond the scope of safety-rated components.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for planning a path of motion for a robot. In some implementations, a candidate path of movement is determined for each of multiple robots. A swept region, for each of the multiple robots, is determined that the robot would traverse through along its candidate path. At least some of the swept regions for the multiple robots is aggregated to determine amounts of overlap among the swept regions at different locations. Force vectors directed outward from the swept regions are assigned, wherein the force vectors have different magnitudes assigned according to the respective amounts of overlap of the swept regions at the different locations. A path for a particular robot to travel is determined based on the swept regions and the assigned magnitudes of the forces.

Abstract: Provided is a self-powered vehicle, comprising: a mechanical drive system, a set of sensors, and a controller coupled to the mechanical drive system to move the vehicle. The self-powered vehicle can operate in a plurality of modes, including a pair mode and a smart behavior mode. In pair mode the vehicle follows the trajectory of a user and in smart behavior mode the vehicle performs autonomous behavior. The self-powered vehicle operates with hysteresis dynamics, such that the movements of the vehicle are consistent with ergonomic comfort of the user and third-party pedestrian courtesy. The self-powered vehicle can operate with other self-powered vehicles in a convoy.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controlling robotic movements. One of the methods includes receiving, for a robot, a definition of a plurality of sensor-based skills to be executed in sequence, wherein each skill is associated with an entry point and an exit point; generating a motion plan for the robot, including: generating, for a first skill of the plurality of sensor-based skills, a first path from a first entry point of the first skill to a second point at which a sensor-based interaction of the first skill begins, and generating, for the first skill of the plurality of sensor-based skills, a second path from a third point at which the sensor-based interaction of the first skill ends to a first exit point of the first skill; and executing the motion plan for the robot.

Abstract: Systems and methods for utilizing interactive Gaussian processes for crowd navigation are provided. In one embodiment, a system for a crowd navigation of a host is provided. The system includes a processor, a statistical module, and a model module. The processor receives sensor data. The statistical module identifies a number of agents in a physical environment based on the sensor data. The statistical module further calculates a set of Gaussian processes. The set of Gaussian processes includes a Gaussian Process for each agent of the number of agents. The statistical module further determines an objective function based on an intent and a flexibility. The model module generates a model of the number of agents by applying the objective function to the set of Gaussian processes. The model includes a convex configuration of the number of agents in the physical environment.

Abstract: Systems and methods for safe drill floor access. An example method includes commencing operation of a processing device to operate a system for maintaining safety of a human worker in a restricted area of a drill floor during drilling operations. The processing device may receive a presence detection signal and output a first stop control command to the equipment to cause the equipment to stop operating. The processing device may receive an entry request signal, output an entry request indicator via an output device to a human driller within a drill rig control center, receive an entry grant input from the human driller granting entry to the human worker to enter the restricted area, and output a second stop control command to the equipment located in the restricted area to cause the equipment to stop operating in response to the received entry grant input.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for planning by work volumes to avoid conflicts. One of the methods includes receiving a process definition graph for a robot that includes action nodes, wherein the action nodes include (1) transition nodes that represent a motion to be taken by the robot from a respective start location to an end location and (2) task nodes that represent a particular task to be performed by the robot at a particular task location. An initial modified process definition graph that ignores one or more conflicts between respective transition nodes as well as one or more conflicts between respective transition nodes and task nodes is generated from the process definition graph. A refined process definition graph that ignores conflicts between transition nodes and recognizes conflicts between transition nodes and task nodes is generated from the initial modified process definition graph.

Abstract: Warehouse automation and methods of handling materials can be used to enhance the safety and efficiencies of warehouse operations. For example, automation systems and methods that make depalletization processes safer and more efficient are described. In some examples, the depalletization systems described herein include multiple work cells/stations and a depalletization robot that traverses the work cells/stations on a rail so the robot can autonomously move between the stations.

Abstract: A control system is provided. A second robot in this control system has a trajectory calculation unit which calculates a trajectory of the second robot so as to avoid a first robot if it is determined that the first robot and the second robot will collide.

Abstract: A system for controlling a user interface of a teleoperated surgical system, the system comprises a first master controller communicatively coupled to the teleoperated surgical system; and a display device communicatively coupled to the teleoperated surgical system and configured to display a graphical user interface; and wherein the first master controller is configured to transmit a first input signal to an interface controller, the first input signal caused by manual manipulation of the first master controller, the interface controller to use the first input signal to update a graphical user interface presented by the display device.

Abstract: A method for improving performance of at least one hardware processor solving a top-k planning problem includes obtaining, in a memory coupled to the at least one processor, a specification of the planning problem in a planning language; obtaining, in a first iteration carried out by the at least one processor, at least one solution to the planning problem; and modifying the planning problem, in the first iteration carried out by the at least one processor, to forbid the at least one solution. The method further includes repeating, by the at least one processor, the obtaining of the at least one solution and the modifying to forbid the at least one solution, for a plurality of additional iterations, after the first iteration, until a desired number, k, of solutions to the planning problem are found or until no further solutions exist, whichever comes first.

Abstract: Described in detail herein is an automated fulfilment system including a computing system programmed to receive requests from disparate sources for physical objects disposed at one or more locations in a facility. The computing system can combine the requests, and group the physical objects in the requests based on object types or expected object locations. Autonomous robot devices can receive instructions from the computing system to retrieve a group of the physical objects and deposit the physical objects in storage containers.

Abstract: An adaptive robot grasp planning technique for bin picking. Workpieces in a bin having random positions and poses are to be grasped by a robot and placed in a goal position and pose. The workpiece shape is analyzed to identify a plurality of robust grasp options, each grasp option having a position and orientation. The workpiece shape is also analyzed to determine a plurality of stable intermediate poses. Each individual workpiece in the bin is evaluated to identity a set of feasible grasps, and the workpiece is moved to the goal pose if such direct movement is possible. If direct movement is not possible, a search problem is formulated, where each stable intermediate pose is a node. The search problem is solved by evaluating the feasibility and optimality of each link between nodes. Feasibility of each link is evaluated in terms of collision avoidance constraints and robot joint motion constraints.

Abstract: A method for visualizing data generated by a robotic device is presented. The method includes displaying an intended path of the robotic device in an environment. The method also includes displaying a first area in the environment identified as drivable for the robotic device. The method further includes receiving an input to identify a second area in the environment as drivable and transmitting the second area to the robotic device.

Abstract: An trajectory information generating unit for generating trajectory information defining a trajectory for which a picking hand picks a work at a first position and arranges the work at a second position, an execution control unit for operating a picking apparatus based on trajectory information generated by the trajectory information generating unit, and an execution time estimating unit for estimating a period of time from when the picking apparatus receives an instruction for starting an operation on a work to a time when the operation of the picking apparatus on the work is ended are included. The trajectory information generating unit may adjust an amount of calculation based on an estimation result of the execution time estimating unit.

Abstract: In a trajectory generation apparatus, position coordinates of an obstacle existing in a motion space of a robot arm is acquired. A hand position at a second time, which is a time next to a first time, is estimated by using a learning result of machine learning, based on the position coordinates of the obstacle, a subject joint state of the robot arm at the first time, and a target joint state of the robot arm. A non-interfering joint state of the robot arm at which the obstacle does not interfere with the robot arm at the second time is searched for by using the hand position as a restriction.

Abstract: A method for perception and fitting for a stair tracker includes receiving sensor data for a robot adjacent to a staircase. For each stair of the staircase, the method includes detecting, at a first time step, an edge of a respective stair of the staircase based on the sensor data. The method also includes determining whether the detected edge is a most likely step edge candidate by comparing the detected edge from the first time step to an alternative detected edge at a second time step, the second time step occurring after the first time step. When the detected edge is the most likely step edge candidate, the method includes defining, by the data processing hardware, a height of the respective stair based on sensor data height about the detected edge. The method also includes generating a staircase model including stairs with respective edges at the respective defined heights.