面向动物机器人刺激器的光伏电池在线监测系统研究与应用

Research and application of photovoltaic cell online monitoring system for animal robot stimulator

电源是动物机器人获得有效刺激的关键要素,应用光伏电池并监测其参数变化可以指导操作者获得最佳的刺激策略,从而有效延长动物机器人的控制时间,本文旨在研究一种面向动物机器人刺激器的光伏电池在线监测系统。该系统主要由电池信息采样电路、多通道神经信号发生器、电源模块和人机交互界面组成,通过多通道信号发生器适时刺激可以对动物机器人进行远程导航控制,同时电池信息采样电路实时采集电池的电压、电流、温度、电量信息,显示在人机交互系统上,并对电池的异常状态进行报警。通过充放电测试获得电池相关参数,并对不同光照强度下的系统续航时间和神经信号发生器的刺激效果进行了测试。结果表明电池电压、电流、电量采样误差分别小于15 mV、5 mA、6 mAh。在78 320 lx的光照强度下,相较于无光伏电池的系统,续航时间延长了148%,在一定程度上解决了动物机器人的续航问题。刺激实验结果表明系统可以控制动物机器人完成左、右转动作,成功率均在80%以上。本系统所获得的参数可以为研究人员优化动物控制策略提供依据。

Power supply plays a key role in ensuring animal robots to obtain effective stimulation. To extending the stimulating time, there is a need to apply photovoltaic cells and monitor their parameter variations, which can help operators to obtain the optimal stimulation strategy. In this paper, an online monitoring system of photovoltaic cells for animal robot stimulators was presented. It was composed of battery information sampling circuit, multi-channel neural signal generator, power module and human-computer interaction interface. When the signal generator was working,remote navigation control of animal robot could be achieved, and the battery voltage, current, temperature and electricity information was collected through the battery information sampling circuit and displayed on the human-computer interaction system in real time. If there was any abnormal status, alarm would be activated. The battery parameters were obtained by charging and discharging test. The battery life under different light intensity and the stimulation effect of neural signal generator were tested. Results showed that the sampling errors of battery voltage, current and electric quantity were less than 15 mV, 5 mA and 6 mAh, respectively. Compared with the system without photovoltaic cells, the battery life was extended by 148% at the light intensity of 78 320 lx, solving the battery life problem to some extent. When animal robot was stimulated with this system, left and right turns could be controlled to complete with the success rate more than 80%. It will help researchers to optimize animal robot control strategies through the parameters obtained in this system.