Decentralized control of insect walking: A simple neural network explains a wide range of behavioral and neurophysiological results
Decentralized control of insect walking: A simple neural network explains a wide range of behavioral and neurophysiological results
复制标题
DOI：
10.1371/journal.pcbi.1007804
复制
发表时间：
2020-04-01
影响因子：
4.3
通讯作者：
Cruse, Holk
中科院分区：
文献类型：
--
作者：
Schilling, Malte;Cruse, Holk
控制昆虫的六条腿在不可预测的环境中行走是一项具有挑战性的任务，因为必须协调许多自由度。解决这个问题的方案通常是基于一个非常有影响力的概念，即（感觉调制）中央模式发生器（CPG）需要控制步行腿的节奏运动。在这里，我们调查一个不同的观点。为此，我们介绍了一个基于传感器的控制器上操作的人工神经元，被施加到（模拟）昆虫机器人需要利用“循环通过世界”允许简化神经计算。我们表明，这样的分散解决方案，导致适应性行为时，面对不确定的环境，我们证明了广泛的行为从来没有在一个单一的系统处理早期的方法。这包括产生脚步模式的能力，例如速度依赖性的“三足动物”、“四足动物”、“五足动物”以及在竹节虫和果蝇中观察到的各种稳定的中间模式。这些模式被发现是稳定的，对干扰，当从各种腿配置。我们的神经元结构很容易允许开始或中断行走，这对于CPG控制的解决方案来说都是困难的。此外，曲线的协商和在跑步机上行走，对单个腿进行各种治疗，以及向后行走和执行短步骤是可能的。这种方法也可以解释通常解释为支持CPG形成行走基础的想法的神经生理学结果，尽管我们的方法不依赖于明确的CPG样结构。然而，对于非常快速的行走，可能需要应用CPG。我们的神经元结构允许精确定位从各种昆虫研究中已知的特定神经元。有趣的是，在昆虫和甲壳类动物中观察到的特定共同特性表明我们的控制器的重要性超出了昆虫领域。作者总结昆虫能够在复杂的环境中行走和攀爬，这需要连续控制至少18个关节。因此，昆虫的表现甚至超过了现代机器人。虽然机器人被建造成复杂的人工系统，但昆虫的行为被认为依赖于相当简单的控制原理的相互作用。两个主要的假设是，昆虫使用-腿之间的协调-离散步态，和-作为协调腿关节的基础-神经元节律发生器。由于这些原则的应用只允许描述有限数量的行为数据，这两个假设在这里都受到了挑战。首先，没有离散的、单独的步态。相反，有一个连续的新兴的腿模式，因为已经知道很长一段时间的竹节虫，最近也证实了果蝇。第二，关于控制不同关节的腿，我们认为，除了非常快的步行，神经元节律发生器是不需要的，但可能会适得其反的计算效率。相反，我们提出了一个分散的，体现神经元结构，利用感觉反馈和内部状态之间的动态切换。该系统解释了大量行为和神经生理学研究以及不同物种的基本方面提供的数据。
Controlling the six legs of an insect walking in an unpredictable environment is a challenging task, as many degrees of freedom have to be coordinated. Solutions proposed to deal with this task are usually based on the highly influential concept that (sensory-modulated) central pattern generators (CPG) are required to control the rhythmic movements of walking legs. Here, we investigate a different view. To this end, we introduce a sensor based controller operating on artificial neurons, being applied to a (simulated) insectoid robot required to exploit the "loop through the world" allowing for simplification of neural computation. We show that such a decentralized solution leads to adaptive behavior when facing uncertain environments which we demonstrate for a broad range of behaviors never dealt with in a single system by earlier approaches. This includes the ability to produce footfall patterns such as velocity dependent "tripod", "tetrapod", "pentapod" as well as various stable intermediate patterns as observed in stick insects and in Drosophila. These patterns are found to be stable against disturbances and when starting from various leg configurations. Our neuronal architecture easily allows for starting or interrupting a walk, all being difficult for CPG controlled solutions. Furthermore, negotiation of curves and walking on a treadmill with various treatments of individual legs is possible as well as backward walking and performing short steps. This approach can as well account for the neurophysiological results usually interpreted to support the idea that CPGs form the basis of walking, although our approach is not relying on explicit CPG-like structures. Application of CPGs may however be required for very fast walking. Our neuronal structure allows to pinpoint specific neurons known from various insect studies. Interestingly, specific common properties observed in both insects and crustaceans suggest a significance of our controller beyond the realm of insects.Author summaryInsects are able to walk and climb in complex environments, which requires continuous control of at least 18 joints. Thereby insects outperform even modern robots. But while robots are built as sophisticated artificial systems, insect behavior is assumed to rely on the interaction of quite simple control principles. Two predominant assumptions are that insects use-for coordination between legs-discrete gaits, and-as a basis to coordinate the joints of a leg-neuronal rhythm generators. As application of these principles allows description of only a limited amount of behavioral data, both assumptions are challenged here. First, there are no discrete, separate gaits. Instead, there is a continuum of emergent leg patterns as has been known since long for stick insects and recently also confirmed for Drosophila. Second, concerning the control of different joints of a leg, we argue that, apart from very fast walking, neuronal rhythm generators are not required, but may rather be counterproductive as concerns computational efficiency. Instead we propose a decentralized, embodied neuronal structure exploiting sensory feedback and dynamic switching between internal states. This system explains data provided by a large amount of behavioral and neurophysiological studies as well as basic aspects of different species.