Robot Control Scheduling for Mixed-Time Application Loops
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As the degree of freedom (DOF) of robots increases, the complexity and time-consuming nature of control algorithms lead to execution times and data transmission times exceeding the control cycle of the robot, causing synchronization issues and inefficiencies in robot control systems.
Innovation Solution
A control method and system where each application registers its protocol, allowing the decision center to extract and encapsulate relevant data, transmit application data packets, and generate component control instructions based on control data from synchronous and asynchronous applications, with timeout judgments and weight-based data processing to ensure timely and efficient component control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous sequential control process is used, then control reliability is maintained, but control cycle time increases and exceeds robot control requirements
Solution Approach 1:
The patent segments the control process into multiple asynchronous tasks that can execute independently. Instead of a single synchronous sequential process, the control algorithm is divided into multiple tasks with different priorities and time requirements, allowing parallel execution and reducing overall control cycle time while maintaining reliability through priority-based scheduling.
Solution Approach 2:
The patent implements dynamic task scheduling where task priorities and execution timing can be adjusted based on real-time system state. This dynamic approach allows the control system to adapt to varying computational loads and time constraints, optimizing the balance between reliability and response time by dynamically allocating resources to critical tasks.
2Manufacturing precision
If control algorithm complexity increases, then control precision improves, but execution time increases beyond control cycle
Solution Approach 1:
The patent divides complex control algorithms into multiple independent tasks that can be executed in parallel. By segmenting the computational workload, the system achieves high control precision through sophisticated algorithms while maintaining fast execution by distributing computations across multiple parallel task streams with different priority levels.
Solution Approach 2:
The patent employs parameter-based task scheduling where computational parameters such as task priority, execution period, and CPU allocation are dynamically adjusted. This allows the system to allocate more computational resources to precision-critical tasks when needed while maintaining real-time performance by adjusting parameters based on system load and time constraints.
3Loss of information
If data transmission time increases, then data completeness improves, but total control time exceeds cycle limit
Solution Approach 1:
The patent segments data transmission into multiple priority-based streams corresponding to different control tasks. Critical data required for high-priority tasks is transmitted first with guaranteed bandwidth, while less critical data is transmitted in parallel or during idle periods. This segmentation ensures data completeness for time-critical operations without delaying the overall control cycle.
Solution Approach 2:
The patent introduces a data management intermediary layer that buffers, prioritizes, and schedules data transmission between sensors/actuators and control tasks. This intermediary optimizes data flow by pre-processing and filtering data before transmission, reducing the actual transmission time while ensuring that all necessary data is delivered complete and accurate to the appropriate tasks.
Data Source
AI summary
A control method and a control system using the same as are provided. The method includes: extracting a corresponding component data from a component data stream based on an application protocol registered by each of a plurality of applications and encapsulating a corresponding application data packet for the application in response to receiving the component data stream in a current control period; transmitting the corresponding application data packet to each of the applications; obtaining control data reported by each of the applications according to data reporting timing corresponding to the application; and generating a component control instruction for controlling each of the components according to the control data reported by each of the applications in the current control period and the application protocol registered by the application. In this manner, the orderly control of a smart devices within a limited control period can be achieved.