Figure 2 - uploaded by Alan F T Winfield
Content may be subject to copyright.
The Triangle of Life. The pivotal moments that span the triangle are: 1) Conception: A new genome is activated, construction of a new organism starts. 2) Delivery: Construction of the new organism is completed. 3) Fertility: The organism becomes ready to conceive offspring.
Source publication
Evolutionary robotics is heading towards fully embodied evolution in real-time and real-space. In this paper we introduce the Triangle of Life, a generic conceptual framework for such systems in which robots can actually reproduce. This framework can be instantiated with different hardware approaches and different reproduction mechanisms, but in al...
Contexts in source publication
Context 1
... proverbial Cycle of Life revolves around birth. We adopt this stance and define the Triangle of Life as shown in Figure 2. This concept of the Triangle is generic, the only signifi- cant assumption we maintain is the genotype-phenotype di- chotomy. ...
Context 2
... new robotic organism is created first at genotype level and is thus seeded by a new piece of genetic code that is created by mutating or recombining existing pieces of code. Birth is therefore the first stage of life, specified as the interval between the moment of activating a newly created genome (circle 1 in Figure 2) and the moment when the robot or- ganism encoded by this genome is completed (circle 2 in Figure 2). In technical terms, this is the period when mor- phogenesis takes place. ...
Context 3
... new robotic organism is created first at genotype level and is thus seeded by a new piece of genetic code that is created by mutating or recombining existing pieces of code. Birth is therefore the first stage of life, specified as the interval between the moment of activating a newly created genome (circle 1 in Figure 2) and the moment when the robot or- ganism encoded by this genome is completed (circle 2 in Figure 2). In technical terms, this is the period when mor- phogenesis takes place. ...
Context 4
... second stage in the Triangle of Life starts when the morphogenesis of a new robot organism is completed (cir- cle 2 in Figure 2) and ends when this organism acquires the skills necessary for living in the given world and be- comes capable of conceiving offspring (circle 3 in Figure 2). This moment of becoming fertile is less easy to define in general than the other two nodes of the triangle. ...
Context 5
... second stage in the Triangle of Life starts when the morphogenesis of a new robot organism is completed (cir- cle 2 in Figure 2) and ends when this organism acquires the skills necessary for living in the given world and be- comes capable of conceiving offspring (circle 3 in Figure 2). This moment of becoming fertile is less easy to define in general than the other two nodes of the triangle. ...
Context 6
... third stage in the Triangle is the period of maturity. It starts when the organism in question becomes fertile (cir- cle 3 in Figure 2) and leads to a new Triangle when this organism conceives a child, i.e., produces a new genome through recombination and/or mutation (circle 1). 2 It should be noted that at this point we switch perspectives: the be- ginning of a new life marks the beginning of another Tri- angle belonging to the new organism encoded by the new piece of genome. As for the 'old' organism nothing needs to end here. ...
Citations
... The approach is neatly broken down into three components: fabrication of a robot from a genotype, learning in the physical world, and finally "mature life" in which tasks are performed, performance is evaluated, and the robot's morphology and/or controller is passed to the next-generation. This cycle has been termed the Triangle of Life (Eiben et al., 2013). Of these three stages, learning is currently the least well developed. ...
While evolutionary robotics can create novel morphologies and controllers that are well-adapted to their environments, learning is still the most efficient way to adapt to changes that occur on shorter time scales. Learning proposals for evolving robots to date have focused on new individuals either learning a controller from scratch, or building on the experience of direct ancestors and/or robots with similar configurations. Here we propose and demonstrate a novel means for social learning of gait patterns, based on sensorimotor synchronization. Using movement patterns of other robots as input can drive nonlinear decentralized controllers such as CPGs into new limit cycles, hence encouraging diversity of movement patterns. Stable autonomous controllers can then be locked in, which we demonstrate using a quasi-Hebbian feedback scheme. We propose that in an ecosystem of robots evolving in a heterogeneous environment, such a scheme may allow for the emergence of generalist task-solvers from a population of specialists.
... In recent years, researchers have introduced the idea of biological evolution into integrated design automation for morphologies and controllers of intelligent robots [123][124][125][126], which can automatically identify the optimal designs of intelligent robots according to fitness functions determined by given tasks or environments. Based on these ideas, some studies [127][128][129] have proposed an underlying system architecture called the triangle of life, which consists of three stages: morphogenesis, infancy, and mature life. This system allows for a population of robotic organisms that evolve and adapt to the given environment. ...
Design automation is a core technology in industrial design software and an important branch of knowledge-worker automation. For example, electronic design automation (EDA) has played an important role in both academia and industry. Design automation for intelligent robots refers to the construction of unified modular graph models for the morphologies (body), controllers (brain), and vision systems (eye) of intelligent robots under digital twin architectures, which effectively supports the automation of the morphology, controller, and vision system design processes of intelligent robots by taking advantage of the powerful capabilities of genetic programming, evolutionary computation, deep learning, reinforcement learning, and causal reasoning in model representation, optimization, perception, decision making, and reasoning. Compared with traditional design methods, MOdular DEsigN Automation (MODENA) methods can significantly improve the design efficiency and performance of robots, effectively avoiding the repetitive trial-and-error processes of traditional design methods, and promoting automatic discovery of innovative designs. Thus, it is of considerable research significance to study MODENA methods for intelligent robots. To this end, this paper provides a systematic and comprehensive overview of applying MODENA in intelligent robots, analyzes the current problems and challenges in the field, and provides an outlook for future research. First, the design automation for the robot morphologies and controllers is reviewed, individually, with automated design of control strategies for swarm robots also discussed, which has emerged as a prominent research focus recently. Next, the integrated design automation of both the morphologies and controllers for robotic systems is presented. Then, the design automation of the vision systems of intelligent robots is summarized when vision systems have become one of the most important modules for intelligent robotic systems. Then, the future research trends of integrated “Body-Brain-Eye” design automation for intelligent robots are discussed. Finally, the common key technologies, research challenges and opportunities in MODENA for intelligent robots are summarized.
... Thus, the joint evolution of morphologies and controllers inherently leads to a potential body-brain mismatch. This problem has been originally noted by (Eiben et al., 2013) and recently revisited in . The proposed solution is the addition of learning. ...
... This three-stage system methodology, dubbed The Triangle of Life, has been described in (Eiben et al., 2013), but to-date there are hardly any studies into the workings of such systems. One reason is that in general there are only a few papers on the joint evolution of morphology and controller, the majority of work in evolutionary robotics considers the evolution of brains within a fixed body. ...
... Cheney et al. (2018) implemented a so-called morphological innovation protection mechanism which allows additional optimization of the controller in a "newborn" body by using mutations of the inherited brain. Technically this is similar to the Infancy stage concept within the Triangle of Life (Eiben et al., 2013). However, there is an important difference: during the protected period a robot can produce offspring (is eligible for parent selection), but it cannot be removed from the population (is exempted from survivor selection), whereas in our infant learning stage the robot is exempted from both selection mechanisms, it cannot produce offspring and it cannot be removed. ...
Simultaneously evolving morphologies (bodies) and controllers (brains) of robots can cause a mismatch between the inherited body and brain in the offspring. To mitigate this problem, the addition of an infant learning period has been proposed relatively long ago by the so-called Triangle of Life approach. However, an empirical assessment is still lacking to-date. In this paper, we investigate the effects of such a learning mechanism from different perspectives. Using extensive simulations we show that learning can greatly increase task performance and reduce the number of generations required to reach a certain fitness level compared to the purely evolutionary approach. Furthermore, we demonstrate that the evolved morphologies will be also different, even though learning only directly affects the controllers. This provides a quantitative demonstration that changes in the brain can induce changes in the body. Finally, we examine the learning delta defined as the performance difference between the inherited and the learned brain, and find that it is growing throughout the evolutionary process. This shows that evolution produces robots with an increasing plasticity, that is, consecutive generations become better learners and, consequently, they perform better at the given task. Moreover, our results demonstrate that the Triangle of Life is not only a concept of theoretical interest, but a system methodology with practical benefits.
... One of the possible reasons is that jointly evolving morphologies and controllers can be very challenging [3]. A key issue noted long ago [7] is that the stochastic nature of reproduction can lead to a body-brain mismatch problem: Even though parents have well-matching bodies and brains, recombination and mutation can shuffle the parental genotypes such that the resulting body and brain combination will not fit well. Consequently, causing sub-optimal behaviour in the offspring. ...
... A generic system architecture to implement this solution is the Triangle of Life framework that integrates evolution and lifetime learning [7]. The essence is to have newborn robots perform a learning process that optimizes their inherited brain quickly after birth. ...
Evolving morphologies and controllers of robots simultaneously leads to a problem: Even if the parents have well-matching bodies and brains, the stochastic recombination can break this match and cause a body-brain mismatch in their offspring. We argue that this can be mitigated by having newborn robots perform a learning process that optimizes their inherited brain quickly after birth. We compare three different algorithms for doing this. To this end, we consider three algorithmic properties, efficiency, efficacy, and the sensitivity to differences in the morphologies of the robots that run the learning process.
... In such a scenario, adaptability is pivotal, since robotic agents are called upon dealing with changing environmental conditions and, possibly, changing mission requirements. Moreover, it is not unlikely that future robotic agents would undergo the same stages of life (birth, maturity, death or disposal) of biological agents [3]. As such, learning and adaptation should inherently guide "infant" robots towards maturity [4], just like in animals. ...
... The baseline controller is different from the Hebbian controller in two ways. First, the weights of the MLP of the former stay the same for the entire simulation, while in the latter they change at every time step according with Equation (3). Second, in the baseline controller we optimize the weights, differently than in the Hebbian controller, where we optimize the ABCD parameters. ...
According to Hebbian theory, synaptic plasticity is the ability of neurons to strengthen or weaken the synapses among them in response to stimuli. It plays a fundamental role in the processes of learning and memory of biological neural networks. With plasticity, biological agents can adapt on multiple timescales and outclass artificial agents, the majority of which still rely on static Artificial Neural Network (ANN) controllers. In this work, we focus on Voxel-based Soft Robots (VSRs), a class of simulated artificial agents, composed as aggregations of elastic cubic blocks. We propose a Hebbian ANN controller where every synapse is associated with a Hebbian rule that controls the way the weight is adapted during the VSR lifetime. For a given task and morphology, we optimize the controller for the task of locomotion by evolving, rather than the weights, the parameters of the Hebbian rules. Our results show that the Hebbian controller is comparable, often better than a non-Hebbian baseline and that it is more adaptable to damages. We also provide novel insights into the inner workings of plasticity and demonstrate that “true” learning does take place, as the evolved controllers improve over the lifetime and generalize well.
... In such a scenario, adaptability is pivotal, since robotic agents are called upon dealing with changing environmental conditions and, possibly, changing mission requirements. Moreover, it is not unlikely that future robotic agents would undergo the same stages of life (birth, maturity, death or disposal) of biological agents [3]. As such, learning and adaptation should inherently guide "infant" robots towards maturity [4], just like in animals. ...
... The baseline controller is different from the Hebbian controller in two ways. First, the weights of the MLP of the former stay the same for the entire simulation, while in the latter they change at every time step according with Equation (3). Second, in the baseline controller it is precisely the weights that we optimize, differently than in the Hebbian controller, where we optimize the ABCD parameters. ...
div>According to Hebbian theory, synaptic plasticity is the ability of neurons to strengthen or weaken the synapses among them in response to stimuli. It plays a fundamental role in the processes of learning and memory of biological neural networks. With plasticity, biological agents can adapt on multiple timescales and outclass artificial agents, the majority of which still rely on static Artificial Neural Network (ANN) controllers. In this work, we focus on Voxel-based Soft Robots (VSRs), a class of simulated artificial agents, composed as aggregations of elastic cubic blocks. We propose a Hebbian ANN controller where every synapse is associated with a Hebbian rule that controls the way the weight is adapted during the VSR lifetime. For a given task and morphology, we optimize the controller for the task of locomotion by evolving, rather than the weights, the parameters of the Hebbian rules. Our results show that the Hebbian controller is comparable, often better than a non-Hebbian baseline and that it is more adaptable to unforeseen damages. We also provide novel insights into the inner workings of plasticity and demonstrate that "true" learning does take place, as the evolved controllers improve over the lifetime and generalize well.</div
... In such a scenario, adaptability is pivotal, since robotic agents are called upon dealing with changing environmental conditions and, possibly, changing mission requirements. Moreover, it is not unlikely that future robotic agents would undergo the same stages of life (birth, maturity, death or disposal) of biological agents [3]. As such, learning and adaptation should inherently guide "infant" robots towards maturity [4], just like in animals. ...
... The baseline controller is different from the Hebbian controller in two ways. First, the weights of the MLP of the former stay the same for the entire simulation, while in the latter they change at every time step according with Equation (3). Second, in the baseline controller it is precisely the weights that we optimize, differently than in the Hebbian controller, where we optimize the ABCD parameters. ...
According to Hebbian theory, synaptic plasticity is the ability of neurons to strengthen or weaken the synapses among them in response to stimuli. It plays a fundamental role in the processes of learning and memory of biological neural networks. With plasticity, biological agents can adapt on multiple timescales and outclass artificial agents, the majority of which still rely on static Artificial Neural Network (ANN) controllers. In this work, we focus on Voxel-based Soft Robots (VSRs), a class of simulated artificial agents, composed as aggregations of elastic cubic blocks. We propose a Hebbian ANN controller where every synapse is associated with a Hebbian rule that controls the way the weight is adapted during the VSR lifetime. For a given task and morphology, we optimize the controller for the task of locomotion by evolving, rather than the weights, the parameters of the Hebbian rules. Our results show that the Hebbian controller is comparable, often better than a non-Hebbian baseline and that it is more adaptable to unforeseen damages. We also provide novel insights into the inner workings of plasticity and demonstrate that "true" learning does take place, as the evolved controllers improve over the lifetime and generalize well.
... Thus, the joint evolution of morphologies and controllers inherently leads to a potential body-brain mismatch. This problem has been originally noted by [Eiben et al., 2013] and recently revisited in . The proposed solution is the addition of learning. ...
... In this paper, we use the Triangle of Life framework to integrate evolution and life time learning [Eiben et al., 2013]. The essence is to have newborn robots perform a learning process that optimizes their inherited brain quickly after birth. ...
Simultaneously evolving morphologies (bodies) and controllers (brains) of robots can cause a mismatch between the inherited body and brain in the offspring. To mitigate this problem, the addition of an infant learning period by the so-called Triangle of Life framework has been proposed relatively long ago. However, an empirical assessment is still lacking to-date. In this paper we investigate the effects of such a learning mechanism from different perspectives. Using extensive simulations we show that learning can greatly increase task performance and reduce the number of generations required to reach a certain fitness level compared to the purely evolutionary approach. Furthermore, although learning only directly affects the controllers, we demonstrate that the evolved morphologies will be also different. This provides a quantitative demonstration that changes in the brain can induce changes in the body. Finally, we examine the concept of morphological intelligence quantified by the ability of a given body to learn. We observe that the learning delta, the performance difference between the inherited and the learned brain, is growing throughout the evolutionary process. This shows that evolution is producing robots with an increasing plasticity, that is, consecutive generations are becoming better and better learners which in turn makes them better and better at the given task. All in all, our results demonstrate that the Triangle of Life is not only a concept of theoretical interest, but a system architecture with practical benefits.
... 1) Centralized, externalized reproduction. A rigorous way of maintaining control over the system would be to set it up such that robot reproduction cannot take place "in the wild" but only in a centralized infrastructure-a reproduction center-where robot offspring can be made, for instance by 3D-printers and automated assembly facilities (Eiben et al., 2013;Hale et al., 2019). Limiting the reproduction to a single or a few centers not only allows keeping track of robot numbers, but also provides the option to restrict the number of robots produced per day. ...
Rapid developments in evolutionary computation, robotics, 3D-printing, and material science are enabling advanced systems of robots that can autonomously reproduce and evolve. The emerging technology of robot evolution challenges existing AI ethics because the inherent adaptivity, stochasticity, and complexity of evolutionary systems severely weaken human control and induce new types of hazards. In this paper we address the question how robot evolution can be responsibly controlled to avoid safety risks. We discuss risks related to robot multiplication, maladaptation, and domination and suggest solutions for meaningful human control. Such concerns may seem far-fetched now, however, we posit that awareness must be created before the technology becomes mature.
... This is not surprising, considering that the joint evolution of morphologies and controllers implies two search spaces and the search space for the brain changes with every new robot body produced. Evolving morphologies and controllers of robots simultaneously leads to a problem which has been noted long ago, being the body-brain mismatch problem [1]: Even though parents have well-matching bodies and brains, recombination and mutation can shuffle the parental genotypes such that the resulting body and brain combination might not fit well. Consequently, causing sub-optimal behaviour in the offspring. ...
... In this paper, we use the Triangle of Life framework ( Figure 1) to integrates evolution and life time learning [1]. The essence is to have newborn robots perform a learning process that optimizes their inherited brain quickly after birth. ...
When controllers (brains) and morphologies (bodies) of robots simultaneously evolve, this can lead to a problem, namely the brain & body mismatch problem. In this research, we propose a solution of lifetime learning. We set up a system where modular robots can create offspring that inherit the bodies of parents by recombination and mutation. With regards to the brains of the offspring, we use two methods to create them. The first one entails solely evolution which means the brain of a robot child is inherited from its parents. The second approach is evolution plus learning which means the brain of a child is inherited as well, but additionally is developed by a learning algorithm - RevDEknn. We compare these two methods by running experiments in a simulator called Revolve and use efficiency, efficacy, and the morphology intelligence of the robots for the comparison. The experiments show that the evolution plus learning method does not only lead to a higher fitness level, but also to more morphologically evolving robots. This constitutes a quantitative demonstration that changes in the brain can induce changes in the body, leading to the concept of morphological intelligence, which is quantified by the learning delta, meaning the ability of a morphology to facilitate learning.
