Robot Maneuver Planning With Iterative Constraint Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic systems face challenges in rapidly authoring and modifying movement routines due to the need for extensive hardcoding, which limits flexibility and can result in failures or damage when not accounting for environmental and physical constraints.
Innovation Solution
A maneuver controller that allows for dynamic planning and execution of robot movements with minimal user input, using a flexible framework that translates user-readable inputs into motion commands, accounting for the robot's current state and environment through a problem generator and solver system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extensive hardcoding is used to program robot movements, then the robot can perform specific tasks, but the time and cost required to author and modify movement routines increases significantly
Solution Approach 1:
The system enables robots to automatically generate and optimize their own movement routines through self-planning algorithms. The robot uses onboard computing resources to autonomously calculate trajectories and control commands, eliminating the need for extensive manual hardcoding and allowing rapid modification of behaviors.
Solution Approach 2:
The invention transforms the approach from fixed hardcoded programs to dynamic parameter-based planning. By changing the control paradigm to use optimization parameters and cost functions, the system can adapt movements to different tasks and environments without requiring complete reprogramming.
2Productivity
If preprogrammed movement routines are used, then the robot can execute tasks, but the system fails to account for environmental and physical constraints leading to failures or damage
Solution Approach 1:
The system continuously monitors the robot's state and environmental conditions, using feedback loops to adjust movement plans in real-time. This closed-loop control ensures that the robot accounts for physical constraints and environmental factors, preventing failures and damage while maintaining task execution.
Solution Approach 2:
The invention transitions from static preprogrammed routines to dynamic real-time planning. The robot continuously recalculates optimal trajectories based on current state and environmental conditions, enabling adaptive response to changing circumstances while maintaining productivity.
3Reliability
If detailed hardcoding is implemented to account for all constraints, then movement safety improves, but the complexity and cost of programming increases
Solution Approach 1:
The robot autonomously handles the complexity of constraint satisfaction through self-planning algorithms. Instead of requiring programmers to manually encode all safety constraints, the system automatically calculates safe trajectories by considering physical limits and environmental factors, reducing programming complexity while maintaining safety.
Solution Approach 2:
The invention replaces manual programming mechanisms with automated computational planning. Complex constraint satisfaction that would require extensive hardcoding is substituted with real-time optimization algorithms, reducing the mechanical burden of detailed programming while ensuring movement safety.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
A dynamic planning controller (200) receives a maneuver (210) and a current state (202) and transforms the maneuver and the current state into a nonlinear optimization problem (222). The nonlinear optimization problem is configured to optimize an unknown force and an unknown position vector. At a first time instance (Ii, to), the controller linearizes the nonlinear optimization problem into a first linear optimization problem and determines a first solution (232) to the first linear optimization problem using quadratic programming. At a second time instance (h, ti), the controller linearizes the nonlinear optimization problem into a second linear optimization problem based on the first solution and determines a second solution to the second linear optimization problem based on the first solution using the quadratic programming. The controller also generates a joint command (204) to control motion of the robot during the maneuver based on the second solution.