Quadruped Motion Research Articles

NeuroDog

Virtual reality (VR) allows us to immerse ourselves in alternative worlds in which we can embody avatars to take on new identities. Usually, these avatars are humanoid or possess very strong anthropomorphic qualities. Allowing users of VR to embody non-humanoid virtual characters or animals presents additional challenges. Extreme morphological differences and the complexities of different characters' motions can make the construction of a real-time mapping between input human motion and target character motion a difficult challenge. Previous animal embodiment work has focused on direct mapping of human motion to the target animal via inverse kinematics. This can lead to the target animal moving in a way which is inappropriate or unnatural for the animal type. We present a novel real-time method, incorporating two neural networks, for mapping human motion to realistic quadruped motion. Crucially, the output quadruped motions are realistic, while also being faithful to the input user motions. We incorporate our mapping into a VR embodiment system in which users can embody a virtual quadruped from a first person perspective. Further, we evaluate our system via a perceptual experiment in which we investigate the quality of the synthesised motion, the system's response to user input and the sense of embodiment experienced by users. The main findings of the study are that the system responds as well as traditional embodiment systems to user input, produces higher quality motion and users experience a higher sense of body ownership when compared to a baseline method in which the human to quadruped motion mapping relies solely on inverse kinematics. Finally, our embodiment system relies solely on consumer-grade hardware, thus making it appropriate for use in applications such as VR gaming or VR social platforms.

Research on high-speed bounding gait analysis and control of quadruped robot

In order to improve the performance of the quadruped robot, the impact of high-speed gait is an important target to measure the performance of the robot. This paper elaborates the research methods of high-speed gait movement of quadrupeds, and a multi-rigid-body dynamics modelling method is presented. According to the analysis of high-speed bounding gait, the dynamic modelling of quadruped motion system is calculated. The motion planning for a rigid-torso quadruped robot based on Poincare Map is presented, and the high-speed bounding gait motion control strategy is established. The simulation experiments have been carried out, and the results have demonstrated the effectiveness of the motion planning model and high-speed bounding method.

Optimization of Central Pattern Generator based Quadruped Animation using Time Series with Genetic Algorithm

Generating believable quadruped motion is a challenging task for application like Games, Virtual Reality and Augmented Reality, where runtime user interaction is needed. In order to successfully generate the believable motion, the runtime or dynamic adjustment to motion gaits is essential. As the use of Central Pattern Generator (CPG) is vital in generating range of dynamic quadruped motions, the optimization is crucial for smooth motion curves. We use the Genetic Algorithm (GA) based tuning technique to optimize the motion curves generated through CPG. The quadruped motion generated through CPG is applied on Skeletal joints and then optimized through fitting the artificial motion parameters tuned using genetic algorithm. The results generated shows much smooth and stable quadruped motion with believable gait patterns.

Mode-adaptive neural networks for quadruped motion control

Quadruped motion includes a wide variation of gaits such as walk, pace, trot and canter, and actions such as jumping, sitting, turning and idling. Applying existing data-driven character control frameworks to such data requires a significant amount of data preprocessing such as motion labeling and alignment. In this paper, we propose a novel neural network architecture called Mode-Adaptive Neural Networks for controlling quadruped characters. The system is composed of the motion prediction network and the gating network. At each frame, the motion prediction network computes the character state in the current frame given the state in the previous frame and the user-provided control signals. The gating network dynamically updates the weights of the motion prediction network by selecting and blending what we call the expert weights, each of which specializes in a particular movement. Due to the increased flexibility, the system can learn consistent expert weights across a wide range of non-periodic/periodic actions, from unstructured motion capture data, in an end-to-end fashion. In addition, the users are released from performing complex labeling of phases in different gaits. We show that this architecture is suitable for encoding the multi-modality of quadruped locomotion and synthesizing responsive motion in real-time.

Onset of pup locomotion coincides with loss of NR2C/D-mediated cortico-striatal EPSCs and dampening of striatal network immature activity

Adult motor coordination requires strong coincident cortical excitatory input to hyperpolarized medium spiny neurons (MSNs), the dominant neuronal population of the striatum. However, cortical and subcortical neurons generate during development large ongoing patterns required for activity-dependent construction of networks. This raises the question of whether immature MSNs have adult features from early stages or whether they generate immature patterns that are timely silenced to enable locomotion. Using a wide range of techniques including dynamic two-photon imaging, whole cell or single-channel patch clamp recording in slices from Nkx2.1-GFP mice, we now report a silencing of MSNs that timely coincides with locomotion. At embryonic stage (as early as E16) and during early postnatal days, genetically identified MSNs have a depolarized resting membrane potential, a high input resistance and lack both inward rectifying (IKIR) and early slowly inactivating (ID) potassium currents. They generate intrinsic voltage-gated clustered calcium activity without synaptic components. From postnatal days 5–7, the striatal network transiently generates synapse-driven giant depolarizing potentials when activation of cortical inputs evokes long lasting EPSCs in MSNs. Both are mediated by NR2C/D-receptors. These immature features are abruptly replaced by adult ones before P10: MSNs express IKIR and ID and generate short lasting, time-locked cortico-striatal AMPA/NMDA EPSCs with no NR2C/D component. This shift parallels the onset of quadruped motion by the pup. Therefore, MSNs generate immature patterns that are timely shut off to enable the coordination of motor programs.

Animating Quadrupeds: Methods and Applications

Abstract Films like Shrek, Madagascar, The Chronicles of Narnia and Charlotte's web all have something in common: realistic quadruped animations. While the animation of animals has been popular for a long time, the technical challenges associated with creating highly realistic, computer generated creatures have been receiving increasing attention recently. The entertainment, education and medical industries have increased the demand for simulation of realistic animals in the computer graphics area. In order to achieve this, several challenges need to be overcome: gathering and processing data that embodies the natural motion of an animal – which is made more difficult by the fact that most animals cannot be easily motion‐captured; building accurate kinematic models for animals, with adapted animation skeletons in particular; and developing either kinematic or physically‐based animation methods, either by embedding some a priori knowledge about the way that quadrupeds locomote and/or adopting examples of real motion. In this paper, we present an overview of the common techniques used to date for realistic quadruped animation. This includes an outline of the various ways that realistic quadruped motion can be achieved, through video‐based acquisition, physics based models, inverse kinematics or some combination of the above.

Gait Patterns of Quadrupeds and Natural Vibration Modes

Quadruped animals switch gait patterns with speed for energy-effective movement. This is similar to the phenomenon that excited natural vibration modes switch with vibration frequency in a multi-degree-of-freedom system. Therefore, in this paper, it is assumed that quadruped animals move by using the natural vibration of their own musculoskeletal systems. In the simplest rigid-body-link model consisting of one body and four legs, there are natural vibration modes similar to the gait patterns (trot, pace, and gallop) of quadruped animals. However, all the natural frequencies in the model exist near the natural frequency of the free leg and are accordingly different from the walking frequencies of actual quadruped animals. When a scapula and a pelvis are added to the rigid-body-link model on the basis of observations of quadruped motion, the natural frequency of the gallop mode used at high speed increases greatly and approaches the walking frequency. If the body characteristics of a horse are applied to the rigid-body-link model with leg joints, the natural vibration modes of the model are close to the gait patterns of the horse.

Control Strategy for Dynamic Locomotion Design of Walking Quadruped

The purpose of this paper is to design the control for a dynamic locomotion of a quadruped. To reach this goal some control strategy is proposed. We simulate a quadruped walking on a horizontal plane. Our walking robot is composed of a body (platform) and four identical two-link legs. We examine a quadruped gait which is similar to gallop, but without fly. This gait involves simultaneous motions of two front and two rear legs. By this way the quadruped model with four legs is reduced to the model with two virtual legs. We consider the quadruped motion in sagittal plane. This motion is divided into four phases: a single support phase on the rear virtual leg, a double support phase, a single support phase on the front virtual leg and a double support phase finally. Quadruped transfers from the double support phase to the phase with supporting rear virtual leg by pushing the front leg off the ground. During this single support phase the distance between the feet increases. After, the front leg touches the ground with passive impact and double support phase begins. Then quadruped transfers from this double support phase to the phase with supporting front leg by pushing the rear leg off the ground. During this single support phase the distance between the feet decreases. After, the rear leg touches the ground with passive impact and next double support phase begins. By this way, quadruped moves forward. For the realization of proposed here control strategy we use technic of computed control torques.