Parallel Inverse Kinematics Hardware for Real-Time Robot Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-degree-of-freedom robots face challenges in real-time control due to slow calculation speed and high energy consumption in inverse kinematics solutions, particularly with CPU-based calculations being time-consuming and GPU-based approaches being costly and power-intensive.
Innovation Solution
An inverse kinematics accelerator system with high parallelism and low power consumption, comprising a parameter initialization module, inverse kinematics scheduler, Jacobian calculation unit, pose update units, and parameter selector, utilizing a dedicated hardware system to quickly calculate joint angles for robots with multiple degrees of freedom.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If CPU-based calculation is used for inverse kinematics, then the robot can handle high-degree-of-freedom joints, but the calculation speed is too slow to satisfy real-time control requirements
Solution Approach 1:
The patent replaces the general-purpose CPU computational system with a dedicated parallel hardware accelerator specifically designed for inverse kinematics calculations. This hardware substitution enables real-time control by performing matrix operations and iterative solving in parallel, achieving calculation speeds that satisfy real-time requirements while supporting high-degree-of-freedom robots.
Solution Approach 2:
The inverse kinematics calculation is divided into multiple parallel processing units that can simultaneously perform different stages of computation (forward kinematics, Jacobian calculation, iterative solving). This segmentation of the computational process into parallel tasks enables the system to handle high-degree-of-freedom calculations at real-time speeds.
2Speed
If GPU-based approach is used for inverse kinematics, then the calculation speed improves, but the cost and power consumption become too high for practical scenarios
Solution Approach 1:
The patent replaces the GPU-based parallel computing system with a custom-designed hardware accelerator that is more energy-efficient for the specific inverse kinematics workload. This dedicated hardware performs the required matrix operations and iterative calculations with lower power consumption than general-purpose GPUs, making it suitable for mobile robots with limited battery power.
Solution Approach 2:
The hardware accelerator is optimized specifically for inverse kinematics calculations rather than general-purpose computing. This localized optimization means the hardware is highly efficient at the specific tasks required (matrix multiplication, iterative solving) while consuming less power than a GPU would for the same specialized workload.
3Ease of manufacture
If CPU calculation is used for inverse kinematics, then the system is simple to implement, but the calculation consumes too much time for real-time control
Solution Approach 1:
The patent replaces the software-based CPU calculation approach with a hardware-based accelerator. While the hardware design is more complex than software, once implemented, it provides deterministic real-time performance that software cannot match. The hardware accelerator performs all inverse kinematics calculations in parallel, reducing calculation time to meet real-time control requirements.
4Adaptability or versatility
If high-degree-of-freedom joints are added to increase robot capability, then the robot can accomplish more tasks, but the inverse kinematics calculation becomes more difficult and slower
Solution Approach 1:
The patent segments the inverse kinematics calculation into modular parallel processing units, each handling specific degrees of freedom or calculation stages. This modular segmentation allows the system to scale to high-degree-of-freedom robots without proportionally increasing calculation complexity, as each segment can be processed independently and in parallel.
Solution Approach 2:
The patent transitions from sequential calculation (single dimension of time) to parallel calculation (adding the dimension of concurrent processing). By implementing multiple parallel processing units that simultaneously handle different aspects of the inverse kinematics problem, the system can accommodate high-degree-of-freedom robots without increasing the time complexity of the calculation.
Data Source
AI summary
An inverse kinematics solution system for use with a robot, which is used for obtaining a joint angle value corresponding to a target pose value on the basis of an inputted target pose value and degree of freedom of a robot and which comprises: a parameters initialization module, an inverse kinematics scheduler, a Jacobian calculating unit, a pose updating unit and a parameters selector. The system is implemented by means of hardware and may quickly obtain motion parameters, which are used for controlling a robot, while reducing power consumption.