Manipulator Control State Machine for Position-Force Transitions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Classic control state machines for manipulators, such as industrial robots, face challenges in seamlessly blending between position- and force-controlled sections, leading to cycle time losses and difficulties in synchronizing filters and observers, due to their deterministic and rigid implementation.

Innovation Solution

Implementing a state machine architecture where commands pass through pre-run, active, and post-run states, allowing control components to settle and synchronize in advance, enabling parallel processing and maintaining a well-defined state for orderly transitions, with additional states for safety and error handling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If classic control state machines run through individual states in a determined sequence with stationary exact stops, then the control structure is simple and deterministic, but cycle time losses occur and transitions between position- and force-controlled sections are difficult

Engineering Contradiction:
Improvecycle timeVSAvoidcontrol structure complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The control method segments the execution of a command into multiple distinct phases: a pre-run phase for initialization and filtering, an active phase for execution, and a post-run phase for finalization. This segmentation allows parallel processing of multiple commands at different phases, reducing cycle time losses while maintaining a manageable control structure through structured state management.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The pre-run phase performs preliminary actions by initializing control components and allowing filters and observers to settle before the command enters the active phase. This preliminary preparation enables smoother transitions between control modes and eliminates waiting time during state transitions, directly addressing the cycle time loss problem.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If filters, observers and parameter identifications need lead time to settle down, then measurement precision is improved, but cycle time losses occur due to waiting for settlement

Engineering Contradiction:
Improvefilter settlement qualityVSAvoidcycle time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The pre-run phase is specifically designed to provide the necessary lead time for filters, observers, and parameter identifications to settle down before the command becomes active. By performing this settlement process in advance during the pre-run phase, the system ensures measurement precision is achieved without causing cycle time losses during the active execution phase.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The control system dynamically transitions commands between pre-run and active phases, allowing flexible timing for filter settlement. Multiple commands can be at different stages simultaneously, with some in pre-run allowing filters to settle while others are actively executing, thus optimizing both measurement precision and cycle time utilization.

Inventive Principle:
Principle #15Dynamics

3Adaptability or versatility

If a rigid deterministic implementation is used, then the control structure is simple, but adaptability for connecting external processes or modifying control is difficult

Engineering Contradiction:
Improvecontrol modification capabilityVSAvoidcontrol implementation complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The state machine implementation is made dynamic by allowing commands to transition between pre-run and active phases based on readiness conditions rather than fixed deterministic sequencing. This dynamic approach enables external processes to trigger command activation and allows flexible modification of control flow while maintaining a relatively simple state machine structure through the use of well-defined state transitions.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system allows parameter changes and external process connections by modifying the control state from pre-run to active based on external triggers or conditions. This parameter change approach enables adaptability without requiring fundamental changes to the control structure, as the same state machine framework handles both deterministic and externally-triggered command execution.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP2324965B2Method and device for controlling manipulators
Publication Date: 2019.04.17 KUKA DEUT GMBH
  • EP2324965B2 patent drawingFigure 1

AI summary

The method involves processing a set of control commands e.g. hold pose using model-based regulation with observer control command (B), impose a desired force using force regulation control command (F) and travel to desired pose under position regulation control command (P). The commands are executed in a state machine in an active state (A). Each of the commands is executed in the state machine in a preliminary state that is placed before the active state and/or a post-operational state that is placed after the active state, and the set of commands are processed simultaneously. Independent claims are also included for the following: (1) a control device for controlling a manipulator comprising a device for executing a method for controlling a manipulator (2) a computer program for executing a method for controlling a manipulator (3) a computer program product comprising a program code for executing a method for controlling a manipulator.