Bipedal locomotion requires precise rhythm and balance. Here we demonstrate two fitness-function enhancements applied to OpenAI’s 3D Humanoid-v1 walking task using a replica of Salimans et al. ’s evolution strategy (Salimans et al., 2017). The first enhancement reduces control cost, following a start-up period, and the second enhancement penalises poor balance. Individually, each enhancement results in improved gaits and doubles both median speed and median distance walked. Combining the two enhancements results in little further improvement in the absence of noise but is shown to produce gaits that are much more robust to noise in their actions, with median speed, distance and time two to five times those of the default and individual-enhancement gaits at an intermediate noise level.

A habit is formed through the repeated enactment of sensorimotor regularities created and maintained by means of plastic changes on the mechanism that brings them about. This precarious, self-maintaining sensorimotor organization is known as sensorimotor autonomy. One can imagine how some habits would be better suited to the maintenance of a biological individual. Evolution can bias the parameters of the plastic medium over which sensorimotor autonomy emerge so as to be beneficial to biological autonomy. In this work, we show that varying some parameters that bring about plastic changes in the behavior-generating medium, different sensorimotor individuals emerge. The simulation consists of a simple robot coupled with a habit-based controller with a random-based exploratory phase in a one-dimensional environment. The results show that, varying the parameters of such a phase, qualitative different habits emerge characterized by static, monotonic and oscillatory behaviors. Quantitative variations of the oscillatory behavior are also shown using the frequencies distribution obtained from the motor time series of the formed habits. The results are interpreted in terms of how the sensorimotor habitat could emerge from the random traversing of the sensorimotor environment. Finally, a comparison between this model and the skin brain thesis is presented.

Soft robots allow for interesting morphological and behavioral designs because they exhibit more degrees of freedom than robots composed of rigid parts. In particular, voxel-based soft robots (VSRs) —aggregations of elastic cubic building blocks—have attracted the interest of Robotics and Artificial Life researchers. VSRs can be controlled by changing the volume of individual blocks: simple, yet effective controllers that do not exploit the feedback of the environment, have been automatically designed by means of Evolutionary Algorithms (EAs). In this work we explore the possibility of evolving sensing controllers in the form of artificial neural networks: we hence allow the robot to sense the environment in which it moves. Although the search space for a sensing controller is larger than its non-sensing counterpart, we show that effective sensing controllers can be evolved which realize interesting locomotion behaviors. We also experimentally investigate the impact of the VSR morphology on the effectiveness of the search and verify that the sensing controllers are indeed able to exploit their sensing ability for better solving the locomotion task.

Echolocation is the process in which an animal produces a sound and recognises characteristics of its surrounding - for instance, the location of surfaces, objects or pray - by listening to the echoes reflected by the environment. Studies on robot echolocation can be found in the literature. Such works adopt active sensors for emitting sounds, and the echoes reflected from the environment are thus analysed to build up a representation of the robot’s surrounding. In this work, we address the usage of robot ego-noise for echolocation. By ego-noise, we mean the auditory noise (sound) that the robot itself is producing while moving due to the frictions in its gears and actuators. Ego-noise is a result not only of the morphological properties of the robot, but also of its interaction with the environment. We adopt a developmental approach in allowing a wheeled robot to learn how to anticipate characteristics of the environment before actually perceiving them. We programmed the robot to explore the environment in order to acquire the necessary sensorimotor information to learn the mapping between ego-noise, motor, and proximity data. Forward models trained with these data are used to anticipate proximity information and thus to classify whether a specific ego-noise is resulting from the robot being close to or distant from a wall. This experiment shows another promising application of predictive processes, that is for echolocation in mobile robots.

Animals interact with their environment softly through interaction of muscles, tendons, and rigid skeleton. By incorporating flexibility, they reduce ground impact forces and improve locomotive efficiency. Flexibility is also beneficial for robotic systems, although it remains challenging to implement. In this paper, we explore the addition of passive flexibility to a quadrupedal animat; we measure the impact of flexibility on both locomotive performance and energy efficiency of movement. Results show that spine and lower limb flexibility can significantly increase distance traveled when compared to an animat with no flexibility. However, replacing passively flexibile joints with actively controlled joints evolves more effective individuals with similar efficiency. Given these results, the number of joints and joint configuration appear to drive performance increases rather than just the addition of passive flexibility.

Agents exhibiting generalized control are capable of solving a theme of related tasks, rather than a specific instance. Here, generalized control pertains to the locomotive capacity of quadrupedal animats, evaluated when climbing over walls of varying height to reach a target. In prior work, we showed that Lexicase selection is more effective than other evolutionary algorithms for this wall crossing task. Generalized controllers capable of crossing the majority of wall heights are discovered, even though Lexicase selection does not sample all possible environments per generation. In this work, we further constrain environmental sampling during evolution, examining the resilience of Lexicase to the impoverished conditions. Through restricting the range of samples at given points in time as well as fixing environmental exposure over fractions of evolutionary time, we attempt to increase the ‘adjacency’ of environmental samples, and report on the response of the Lexicase algorithm to the pressure of this reduced environmental diversity. Results indicate that Lexicase is robust, producing viable agents even in considerably challenging conditions. We also see a positive correlation between the number of tiebreak events that occur and the success of individuals in a population, except in the most limiting conditions. We argue that the increased number of tiebreaks is a response to local maxima, and the increased diversity resulting from random selection at this point, is a key driver of the resilience of the Lexicase algorithm. We also show that in extreme cases, this relationship breaks down. We conclude that tiebreaking is an important control mechanism in Lexicase operation, and that the breakdown in performance observed in extreme conditions indicates an inability of the tiebreak mechanism to function effectively where population diversity is unable to reflect environmental diversity.