Self-supervised learning of the biologically-inspired obstacle avoidance of hexapod walking robot

  title={Self-supervised learning of the biologically-inspired obstacle avoidance of hexapod walking robot},
  author={Petr {\vC}{\'i}{\vz}ek and Jan Faigl},
  journal={Bioinspiration \& Biomimetics},
In this paper, we propose an integrated biologically inspired visual collision avoidance approach that is deployed on a real hexapod walking robot. The proposed approach is based on the Lobula giant movement detector (LGMD), a neural network for looming stimuli detection that can be found in visual pathways of insects, such as locusts. Although a superior performance of the LGMD in the detection of intercepting objects has been shown in many collision avoiding scenarios, its direct integration… 

Improved Collision Perception Neuronal System Model With Adaptive Inhibition Mechanism and Evolutionary Learning

An improved neuronal system model, called LGMD+, that is featured by novel modelling of spatiotemporal inhibition dynamics with biological plausibilities including 1) lateral inhibitions with global biases defined by a variant of Gaussian distribution, spatially, and 2) an adaptive feed-forward inhibition mediation pathway, temporally is proposed.

High-Fidelity Modeling of Hexapod Walking Robot Locomotion

High-fidelity simulation of mobile robotic platforms is a crucial tool to enable verification of control strategies and algorithms prior their deployment on the real robots. In this thesis, a design

Multi-Target Coordinated Search Algorithm for Swarm Robotics Considering Practical Constraints

The proposed multi-target search strategy for swarm robots is more stable, efficient, and practical than the previous methods, and a distributed neighborhood communication mechanism based on a time-varying characteristic swarm with a restricted random line of sight is proposed.

Product Tests in Virtual Reality: Lessons Learned during Collision Avoidance Development for Drones

: Virtual reality (VR) and real-world simulations have become an important tool for product development, product design, and product tests. Product tests in VR have many advantages, such as


  • V. Roganov
  • Computer Science
    XXI Century: Resumes of the Past and Challenges of the Present plus
  • 2021

Robustness of Bio-Inspired Visual Systems for Collision Prediction in Critical Robot Traffic

This study well complements previous experimentation on such bio-inspired computations for collision prediction in more critical physical scenarios, and for the first time demonstrates the robustness of LGMDs inspired visual systems in critical traffic towards a reliable collision alert system under constrained computation power.



Bio-inspired Collision Detector with Enhanced Selectivity for Ground Robotic Vision System

A novel collision selective visual neural network inspired by LGMD2 neurons in the juvenile locusts is proposed, enhancing the collision selectivity in a bio-inspired way, via constructing a computing ef ficient visual sensor, and realizing the revealed specific characteristic sofLGMD2.

Neural based obstacle avoidance with CPG controlled hexapod walking robot

In this work, we are proposing a collision avoidance system for a hexapod crawling robot based on the detection of intercepting objects using the Lobula giant movement detector (LGMD) connected

Locomotion Control and Gait Planning of a Novel Hexapod Robot Using Biomimetic Neurons

The proposed biomimetic method inspired by central pattern generator (CPG) for locomotion control and gait planning of a novel hexapod robot whose legs can radially free distributed around the robot body succeeds in enabling a hexAPod robot to walk steadily on rugged terrain.

A Bio-inspired Collision Detector for Small Quadcopter

A novel bio-inspired collision detector inspired from Lobula giant movement detector neurons in the locusts is modeled into an STM32F407 Microcontroller Unit (MCU) and its application on a quadcopter is studied.

Modeling Proprioceptive Sensing for Locomotion Control of Hexapod Walking Robot in Robotic Simulator

This work focuses on enhancing a simplified model of the multi-legged robot employed in a realistic robotic simulator to provide high-fidelity proprioceptive sensor signals, based on parameter identification and static and dynamic modeling of the robot.

Differential Evolution and Bayesian Optimisation for Hyper-Parameter Selection in Mixed-Signal Neuromorphic Circuits Applied to UAV Obstacle Avoidance

A neuromorphic spiking neural network model of the LGMD driven by the output of a neuromorphic Dynamic Vision Sensor is presented, which has been optimised to produce robust and reliable responses in the face of the constraints and variability of its mixed signal analogue-digital circuits.

Collision detection in complex dynamic scenes using an LGMD-based visual neural network with feature enhancement

An LGMD-based neural network is proposed with a new feature enhancement mechanism to enhance the expanded edges of colliding objects via grouped excitation for collision detection with complex backgrounds by filtering out isolated excitation caused by background detail.

Adaptive and Energy Efficient Walking in a Hexapod Robot Under Neuromechanical Control and Sensorimotor Learning

The tight combination of neural control with tunable muscle-like functions, guided by sensory feedback and coupled with sensorimotor learning, is a way forward to better understand and solve adaptive coordination problems in multilegged locomotion.

Tactile sensing with servo drives feedback only for blind hexapod walking robot

  • Jakub MrvaJ. Faigl
  • Engineering
    2015 10th International Workshop on Robot Motion and Control (RoMoCo)
  • 2015
The main idea of the proposed approach is to consider only the feedback from the intelligent servo drives to detect the contact point of a leg with the surface to enable a smooth motion of the technically blind robot in rough terrains of various difficulty.

Collision avoidance using a model of the locust LGMD neuron