Continuous-Time Filter Discretization for Multi-Rate Robotic Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Surgical robotic systems face challenges in designing filters for different signal and data sampling rates across various modes of operation, leading to difficulties in implementing appropriate filters due to differing control rates and hardware limitations.
Innovation Solution
A scalable filtering infrastructure is developed, where filters are defined in continuous time and discretized for different control rates, allowing for a library of filters to be generated and deployed as discrete time realizations, enabling convenient generation of filters for distributed systems with multiple control rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If filters are designed for each specific control rate separately, then filter performance for that control rate is optimized, but design time and complexity increase significantly
Solution Approach 1:
The patent transforms filter design from discrete-time domain to continuous-time domain by changing the parameter representation. Continuous-time filters are defined with parameters (cut-off frequency, order) that are independent of sampling rate, then converted to discrete-time implementation at runtime. This parameter transformation resolves the contradiction by enabling single-filter-design-for-multiple-rates.
Solution Approach 2:
The patent creates a universal filter library where a single continuous-time filter definition can serve multiple control rates. The filter library stores filters in continuous-time domain and provides discretization functionality to adapt the same filter to different sampling rates (e.g., 1kHz, 10kHz, 100Hz). This multi-functionality eliminates the need to redesign filters for each rate.
2Productivity
If a library of filters is created for multiple control rates, then filter generation speed increases, but the complexity of managing and discretizing filters increases
Solution Approach 1:
The patent uses copying by creating a standardized template for continuous-time filters that can be instantiated multiple times with different parameters. The same filter template can be copied and adapted to different control rates through discretization, reducing the need to create entirely new filter designs for each rate.
Solution Approach 2:
The patent introduces continuous-time domain as an intermediary representation layer between filter specifications and discrete-time implementations. This intermediary allows filters to be defined independently of sampling rate, then converted to the required discrete form at runtime, simplifying the management complexity.
3Adaptability or versatility
If distributed system components operate at different control rates, then system adaptability to different operational modes improves, but filter design difficulty increases
Solution Approach 1:
The patent makes the filter implementation dynamic by providing runtime discretization capability. The same continuous-time filter definition can be dynamically converted to discrete-time form adapted to the specific control rate required by each distributed component (e.g., user console at 1kHz, robot at 10kHz). This dynamic adaptation enables system versatility without increasing design complexity.
Data Source
AI summary
For a scalable filtering infrastructure, a library of filters each usable at different control rates is provided by defining filters in a continuous time mode despite eventual use for digital filtering. For implementation, a filter is selected and discretized for the desired control rate. The discretized filter is then deployed as a discrete time realization for convolution. In a distributed system with multiple control rates, the library may be used to more rapidly and conveniently generate the desired filters.