| 研究課題/領域番号 |
21K14131
|
|---|
| 研究機関 | 早稲田大学 |
|---|
研究代表者 |
トモ ティト・プラドノ 早稲田大学, 次世代ロボット研究機構, 次席研究員(研究院講師) (00844193)
|
|---|
| 研究期間 (年度) |
2021-04-01 – 2023-03-31
|
| キーワード | stretchable / 3-axis / tactile sensor |
|---|
| 研究実績の概要 |
In FY2021, we developed the prototype of a distributed, stretchable, 3-axis soft skin sensor, "Stretch uSkin". The sensor's working principle is based on Hall-Effect sensing, where the force can be detected by displacing a small permanent magnet embedded inside the silicone structure floating above MLX90393, a 3D magnetometer chip. However, instead of using a rigid Printed Circuit Board (PCB), we also use stretchable PCBs that were manufactured by Cir-tech and Connectec Japan. These PCBs are made of a stretchable substrate where the wires are stretchable silver pastes with a meander pattern.
Several experiments such as measuring bus resistances due to stretching cycle, distance, and speed were conducted to analyze the performance of each PCB design.
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
Even though the sensor can measure distributed 3-axis force, the maximum stretch is limited to only up to 150% of its original length. It is challenging to cover robot joints with this stretch ratio.
We also found that the bus resistance of one out of two sensor models becomes very high after stretching it more than 50 cycles. Both sensors had corrupted output data, possibly because of the unreliable I2C communication due to this problem. It is impractical considering that robots usually will operate on the order of a million times.
In FY2022, we initially plan to cover robot joints with sensors from FY2021 and evaluate their performance. Because of these problems, we must develop another version of a stretchable tactile sensor using a different approach before we can install it on a robot.
|
| 今後の研究の推進方策 |
In FY2022, we will develop the second version of a stretchable skin sensor. The sensor will use coil wires instead of stretchable silver pastes with a meander pattern. Our target is not only to achieve a stretch ratio of more than 300% but also can provide reliable output.
We will conduct several experiments similar to what we did in FY2021 to evaluate the sensor's characteristics.
Finally, we will use the sensor to cover robot joints. For example, Allegro Hand or Nicebot.
|
| 次年度使用額が生じた理由 |
1. To purchase raw materials and equipment for manufacturing stretch PCB in-house and also conducting experiments. For example, copper wires, silicone rubber, 3D printer materials, reference sensor, etc.
2. To manufacture rigid PCBs with magnetometers (MLX90393) and microcontrollers
3. To purchase tools for documentation such as a camera and PC
|
