Robust Readout of Cross-Point Resistive Arrays via Time-Domain Optimization
Organizations: National Institute of Informatics (NII) and with The Graduate University for Advanced Studies (SOKENDAI), 2-1-2 Hitotsubashi, Chiyoda-ku, Tokyo, 101-8430, Japan · CNRS-AIST JRL (Joint Robotics Laboratory), IRL 3218, National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Umezono, Tsukuba, Ibaraki 305-8568, Japan
Abstract
Cross-point resistive arrays are attractive for high-density tactile sensing because they reduce wiring complexity. However, conventional frequency-based readout methods can become inaccurate in robotic sensing environments, where long interconnects and measurement and environmental noise degrade the observed signals. We propose a time-domain optimization method for robust readout of cross-point resistive arrays. The method determines resistance values by fitting a time-domain observation model to the measured waveform over the interval, rather than relying solely on spectral amplitudes. By exploiting the full waveform, the proposed method improves robustness under noisy observation conditions. In simulations of a 16 x 16 array, the proposed method reduced the mean readout error by 20% compared with the baseline method. Furthermore, validation over 1,000 trials showed that the proposed method achieves more robust estimation than the conventional method, reducing the mean error and standard deviation by 24.6% and 77.6%, respectively. The improvement was also confirmed in cross-point array simulations with resistance distributions emulating tactile sensor responses.
Figures & tables
| Frequency [ Hz ] | 5k | 6k | 7k | 8k | 9k | 10k | 11k | 12k | 13k | 14k | 15k | 16k | 17k | 18k | 19k | 20k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| FFT(w/o noise) [ % ] | 5.18 | 5.41 | 4.28 | 7.65 | 4.42 | 5.75 | 1.16 | 1.15 | 1.54 | 1.55 | 1.67 | 1.76 | 2.09 | 2.04 | 2.45 | 2.61 |
| Opt(w/o noise) [ % ] | 1.13 | 3.03 | 7.91 | 1.05 | 4.44 | 2.03 | 9.07 | 1.20 | 1.85 | 2.05 | 2.32 | 2.65 | 3.23 | 3.42 | 4.26 | 5.21 |
| FFT(w/ noise) [ % ] | 0.326 | 0.378 | 2.33 | 0.184 | 0.106 | 0.300 | 1.47 | 0.0196 | 0.171 | 0.333 | 0.976 | 0.454 | 0.0301 | 0.127 | 0.535 | 0.261 |
| Opt(w/ noise) [ % ] | 0.112 | 0.400 | 1.56 | 0.225 | 0.173 | 0.301 | 0.0295 | 0.0369 | 0.166 | 0.361 | 1.02 | 0.459 | 0.00822 | 0.178 | 0.601 | 0.288 |
| Experiment | FFT | Opt | ||
|---|---|---|---|---|
| Max. error [%] | Mean error [%] | Max. error [%] | Mean error [%] | |
| w/o noise | 0.00031 | 0.00015 | 0.00059 | 0.00022 |
| w/ noise | 3.63 | 0.40 | 3.26 | 0.32 |
| Condition | FFT | Opt | ||
|---|---|---|---|---|
| Mean [%] | SD [%] | Mean [%] | SD [%] | |
| (b) | 0.62 | 4.00 | 0.37 | 0.45 |
| (c) | 0.57 | 1.70 | 0.43 | 0.38 |
Noise Robustness
Noise robustness studies how to preserve reliable predictions when data, labels, signals, or computing hardware are corrupted.
See all 99 papers