Robot Motion Control for Real-Time Constraint-Safe Target Poses
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motion control methods for robots and virtual avatars are too slow and cannot be used in real-time, especially when intermediate poses are infeasible due to hardware, environmental, or task constraints.
Innovation Solution
A method and control unit that randomize configurations, predict constraint costs, and determine a target configuration that meets the worst constraint, allowing real-time control of an object to achieve a target pose while considering multiple constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If motion planning algorithms and optimization algorithms are used to generate feasible trajectories and make target poses feasible, then the target pose can be reached while satisfying constraints, but the control method is too slow for real-time applications
Solution Approach 1:
The system pre-computes and stores a motion manifold during an offline phase, which contains pre-planned motion paths and their associated constraint information. During real-time operation, the controller only needs to query this pre-computed manifold and select an appropriate trajectory, avoiding the need for slow online optimization calculations while still ensuring constraint satisfaction.
Solution Approach 2:
The motion space is segmented into a discrete motion manifold with multiple pre-computed trajectories. Instead of continuously optimizing trajectories in real-time, the system divides the problem into offline computation (creating the manifold) and online selection (choosing from pre-computed options), enabling real-time control with guaranteed constraint satisfaction.
2Speed
If inverse kinematics controller is used to compute commands for moving end-effector towards target pose, then the control computation is fast, but the method cannot handle infeasible intermediate poses due to hardware, environmental, or task constraints
Solution Approach 1:
The motion manifold acts as an intermediary between the simple inverse kinematics controller and the complex constraint requirements. The manifold pre-encodes feasible motion paths that satisfy all constraints, allowing the fast IK controller to operate on these pre-validated trajectories without directly handling constraint complexity, thus maintaining both speed and reliability.
3Productivity
If teleoperated control is implemented where operator commands target pose in real-time, then the system is highly responsive to operator input, but traditional motion planning methods cannot keep up with the real-time control requirements
Solution Approach 1:
The computationally intensive motion planning is performed in advance during offline operation to create the motion manifold. During teleoperation, the system only performs lightweight queries and selections from this pre-computed manifold, enabling real-time responsiveness to operator commands without the complexity of online motion planning.
Data Source
AI summary
The disclosure relates to a method for controlling motion of an object towards a target pose. The method includes determining a current state of the object; randomizing multiple configurations of the object with regard to the current state of the object; predicting for each configuration of the multiple configurations of the object a constraint cost with regard to each constraint of one or more constraints; determining for each constraint of the one or more constraints, among the predicted constraints costs that have been predicted for the multiple configurations of the object, a predicted constraint cost that meets the constraint the worst; predicting a target configuration of the object using the multiple configurations of the object, the one or more determined constraint costs that have been determined for the one or more constraints, and the target pose; and controlling the object to change its current state towards the predicted target configuration.