Multimodal Representation Learning for Place Recognition Using Deep Hebbian Predictive Coding

Recognising familiar places is a competence required in many engineering applications that interact with the real world such as robot navigation. Combining information from different sensory sources promotes robustness and accuracy of place recognition. However, mismatch in data registration, dimensionality, and timing between modalities remain challenging problems in multisensory place recognition. Spurious data generated by sensor drop-out in multisensory environments is particularly problematic and often resolved through adhoc and brittle solutions. An effective approach to these problems is demonstrated by animals as they gracefully move through the world. Therefore, we take a neuro-ethological approach by adopting self-supervised representation learning based on a neuroscientific model of visual cortex known as predictive coding. We demonstrate how this parsimonious network algorithm which is trained using a local learning rule can be extended to combine visual and tactile sensory cues from a biomimetic robot as it naturally explores a visually aliased environment. The place recognition performance obtained using joint latent representations generated by the network is significantly better than contemporary representation learning techniques. Further, we see evidence of improved robustness at place recognition in face of unimodal sensor drop-out. The proposed multimodal deep predictive coding algorithm presented is also linearly extensible to accommodate more than two sensory modalities, thereby providing an intriguing example of the value of neuro-biologically plausible representation learning for multimodal navigation.

Sacrificing Animals in the Name of Scientific Authority: The Relationship Between Pro-Scientific Mindset and the Lethal Use of Animals in Biomedical Experimentation

The present research investigated how scientific authority increases the lethal use of animals in biomedical experimentation. In two behavioral studies ( N = 151 and 150), participants were required to incrementally administer 12 doses of a toxic chemical to a 53-cm fish (in reality, a biomimetic robot) for research on animal learning. Consistent with the Engaged Followership Theory on obedience, participants placed in a pro-scientific mindset more severely harmed the laboratory animal. In a cross-sectional study ( N = 351), participants in medical fields endorsed a more pro-scientific attitude than those in paramedical fields, which mediated their support for animal experimentation. Drawing on a representative European sample ( N = 31,238), we also confirmed the specificity of this link by controlling for potential demographic and ideological confounds. In a final study ( N = 1,598), instrumental harm was shown as mediating the link between a pro-scientific attitude and support for animal experimentation.

Pipeline Inspection Tests Using a Biomimetic Robot

This paper presents a biomimetic prototype of a mobile robot that can be used to inspect the subdrainage conditions of pipelines located along different highways in Mexico. Computer-aided design tools have been used to size each of the prototype components as inspired by anatomical spider structure. Springs are integrated to generate proper contact pressure against the pipe walls. The robot locomotion system is implemented with adaptable behaviour for the irregularities of pipelines along its journey. The robot prototype is manufactured in 3D printing with the advantage of having its spare parts easily replaceable. Reported results show internal pipe status through a mini video camera on the top of the robot.

Children’s Evaluations of a Therapy Dog and Biomimetic Robot: Influences of Animistic Beliefs and Social Interaction

Social robots are being used increasingly across a range of settings, including in the context of therapeutic interactions with children. While research has shown that interaction with live therapy dogs can be calming and enjoyable for children, it is currently unclear whether social robots can produce similar outcomes. In this study, 11–12-year old children completed a questionnaire about their biophilic beliefs and attitudes to dogs and robots before engaging in two separate free-play, non-goal directed, non-therapeutic sessions with an interactive biomimetic MiRo-E robot and a living therapy dog in a controlled setting. Behavioural observations of social interaction, initiation and reaction behaviours by the child and dog/robot showed that participants spent a similar amount of time engaging in positive social touch with the robot and the dog, but overall more time interacting with the robot. This may be because the robot was more responsive to the children’s initiation behaviours. In self-report, participants significantly preferred the session with the living dog. However, overall enjoyment was high and more positive emotions were reported following interaction with the robot. The more participants attributed mental attributes and animacy to the dog/robot, the more they enjoyed the interactions, demonstrating that participants’ animistic beliefs were an important factor in their evaluations. Levels of social interaction did not correlate with enjoyment, suggesting that the nature of the interaction was less important than pre-existing participant attitudes in producing reported positive outcomes. Although there were some differences in behaviour and evaluations, these preliminary results suggest that MiRo-E provides a useful comparison to therapy dogs and may be a suitable alternative for use in interventions with children.

Group-level patterns emerge from individual speed as revealed by an extremely social robotic fish

Understanding the emergence of collective behaviour has long been a key research focus in the natural sciences. Besides the fundamental role of social interaction rules, a combination of theoretical and empirical work indicates individual speed may be a key process that drives the collective behaviour of animal groups. Socially induced changes in speed by interacting animals make it difficult to isolate the effects of individual speed on group-level behaviours. Here, we tackled this issue by pairing guppies with a biomimetic robot. We used a closed-loop tracking and feedback system to let a robotic fish naturally interact with a live partner in real time, and programmed it to strongly copy and follow its partner's movements while lacking any preferred movement speed or directionality of its own. We show that individual differences in guppies' movement speed were highly repeatable and in turn shaped key collective patterns: a higher individual speed resulted in stronger leadership, lower cohesion, higher alignment and better temporal coordination of the pairs. By combining the strengths of individual-based models and observational work with state-of-the-art robotics, we provide novel evidence that individual speed is a key, fundamental process in the emergence of collective behaviour.

Calculation of the Size of Biomimetics Robots According to Needs and Purpose of Use

<p class="0abstract"><strong><span>- </span></strong><span>In the 4^th industrial revolution era, more countries and companies tend to use specialized robots to replace workers. Especially in a hazardous and high risks working environment, robots that perform human tasks are also humanistic. It’s even the unique solution to ensure life safety. Currently, scientists are much interested in robots that look, act like animals; including biomimetic robot models have capable of flexible mobility and widely application potential. However, it’s difficult to calculate, design and manufacture dynamic robots. In this article, the author will present an overview of dynamic robots and propose new methods and calculation techniques to design and build biomimetic robots. The experimental results show that proposals have contributed to solve those difficulties and being practically applicable to make a biomimetic robot.</span></p>

Group-level patterns emerge from individual speed as revealed by an extremely social robotic fish

Understanding the emergence of collective behaviour has long been a key research focus in the natural sciences. Besides the fundamental role of social interaction rules, a combination of theoretical and empirical work indicates individual speed may be a key process that drives the collective behaviour of animal groups. Socially-induced changes in speed by interacting animals make it difficult to isolate the effects of individual speed on group-level behaviours. Here we tackled this issue by pairing guppies with a biomimetic robot. We used a closed-loop tracking and feedback system to let a robotic fish naturally interact with a live partner in real time, and programmed it to strongly copy and follow its partner’s movements while lacking any preferred movement speed or directionality of its own. We show that individual differences in guppies’ movement speed were highly repeatable and shaped key collective patterns: higher individual speeds resulted in stronger leadership, lower cohesion, higher alignment, and better temporal coordination in the pairs. By combining the strengths of individual-based models and observational work with state-of-the-art robotics, we provide novel evidence that individual speed is a key, fundamental process in the emergence of collective behaviour.