Using Biologically Inspired Features for Face Processing

In this paper, we show that a new set of visual features, derived from a feed-forward model of the primate visual object recognition pathway proposed by Riesenhuber and Poggio (R&P Model) (Nature Neurosci. 2(11):1019–1025, 1999) is capable of matching the performance of some of the best current representations for face identification and facial expression recognition. Previous work has shown that the Riesenhuber and Poggio Model features can achieve a high level of performance on object recognition tasks (Serre, T., et al. in IEEE Comput. Vis. Pattern Recognit. 2:994–1000, 2005). Here we modify the R&P model in order to create a new set of features useful for face identification and expression recognition. Results from tests on the FERET, ORL and AR datasets show that these features are capable of matching and sometimes outperforming other top visual features such as local binary patterns (Ahonen, T., et al. in 8th European Conference on Computer Vision, pp. 469–481, 2004) and histogram of gradient features (Dalal, N., Triggs, B. in International Conference on Computer Vision & Pattern Recognition, pp. 886–893, 2005). Having a model based on shared lower level features, and face and object recognition specific higher level features, is consistent with findings from electrophysiology and functional magnetic resonance imaging experiments. Thus, our model begins to address the complete recognition problem in a biologically plausible way.     

Biologically Inspired Autonomous Rover Control

Robotic missions beyond 2013 will likely be precursors to a manned habitat deployment on Mars. Such missions require robust control systems for long duration activities. Current single rover missions will evolve into deployment of multiple, heterogeneous cooperating robotic colonies. This paper describes the map-making memory and action selection mechanism of BISMARC ( iologically nspired ystem for ap-based utonomous over ontrol) that is currently under development at the Jet Propulsion Laboratory in Pasadena, CA (Huntsberger and Rose, Neutral Networks, 11(7/8):1497–1510). BISMARC is an integrated control system for long duration missions involving robots performing cooperative tasks.     

RHex: A Biologically Inspired Hexapod Runner

RHex is an untethered, compliant leg hexapod robot that travels at better than one body length per second over terrain few other robots can negotiate at all. Inspired by biomechanics insights into arthropod locomotion, RHex uses a clock excited alternating tripod gait to walk and run in a highly maneuverable and robust manner. We present empirical data establishing that RHex exhibits a dynamical (bouncing) gait—its mass center moves in a manner well approximated by trajectories from a Spring Loaded Inverted Pendulum (SLIP)—characteristic of a large and diverse group of running animals, when its central clock, body mass, and leg stiffnesses are appropriately tuned. The SLIP template can function as a useful control guide in developing more complex autonomous locomotion behaviors such as registration via visual servoing, local exploration via visual odometry, obstacle avoidance, and, eventually, global mapping and localization.     

Biologically Inspired Models to Train Neural Networks

A major stumbling block to the successful implementation of neural networks for nonlinear regression models is overtraining. This paper presents two models to combat overtraining, based on the biological concept that the strength of the connection between neurons develops over time. For the problems investigated, the models produce smooth solutions with no sign of overtraining. For the practical Diminishing Returns problem, the Error Constraints Model, because of its mathematical formulation, determines the minimum number of hidden layer neurons. As a result, the present work is important because, if an acceptable level of error can be specified, then the Error Constraints Model can be used to determine the network architecture. Copies of the workbook implementations of all the models presented in this paper may be downloaded from the authors website.     

Biologically Inspired Visual Motion Detection in VLSI

Visual motion detection is a fundament component of vision, and plays a vital role in scene analysis and understanding for behaving organisms. In computer vision, motion detection requires considerable resources to obtain real-time results. Very Large Scale Integration (VLSI) technology offers a convenient substrate upon which both photosensitive elements and motion extracting circuits can be implemented, thus allowing real-time motion detection. This paper presents two approaches for implementing real-time visual motion detection in VLSI. The two approaches mimic the two primary methods found in biological organisms. Insect motion detection employs local correlation and is implemented very close to the photoreceptors. In contrast, primate motion detection is performed in cortex, using spatiotemporally oriented neural filters. The analysis, construction and results of the hardware models of insect and primate visual motion detection are presented.     

Optimal Protraction of a Biologically Inspired Robot Leg

In this paper, protraction movement, namely forward stepping, of a biologically inspired three-joint robot leg is optimized for minimum energy consumption. Trajectory optimization is performed for various initial-final tip point positions of protraction. A modified version of gradient descent based optimal control algorithm is used. The objective function is modified in steps to jump over many unfeasible and inefficient local optima. The optimized trajectories are used to construct a radial basis function neural network (RBFNN) to interpolate for the untrained regions. The results of optimization are compared with the observations of protraction of stick insects. It is concluded that a direct biological imitation of protraction is not energy efficient. A sample protraction of a leg of the Robot-EA308 is demonstrated in guidance of the optimized trajectory. Energy optimal protraction of a robot leg necessitates flexion of the leg, rather than extension as observed in the stick insects.     

改进的生物激励神经网络的机器人路径规划

介绍了基于生物激励神经网络的移动机器人路径规划。机器人的路径生成过程是由神经网络组成动态变化的神经元活性值状态路线图实现的。通过神经元活性值的传播,机器人被吸引到目标点,而同时障碍物使自己处在活性值最低点,起到推开机器人避碰的目的。仿真研究表明该方法生成的由起始点到目标点的路径是连续的、平滑的、避障的,不会陷入U形障碍物,与障碍物形状和所处位置无关,能对快速变化的环境做出迅速反应。但在当前位置邻近位置中具有最大活性值的位置不惟一的情况下,产生路径可能不理想,即到达目标点的避障路径是较长的,而不是最短或者是接近最短的。文中对该不足进行了分析,并提出了改进方法,使生成路径是最短的或是接近最短。对改进方法进行了仿真,实验结果证明该方法是有效的和可行的。     

改进的生物激励神经网络的机器人路径规划

介绍了基于生物激励神经网络的移动机器人路径规划。机器人的路径生成过程是由神经网络组成动态变化的冲经元活性值状态路线图实现的。通过神经元活性值的传播，机器人被吸引到目标点，而同时障碍物使自己处在活性值最低点，起到推开机器人避碰的目的。仿真研究表明该方法生成的由起始点到目标点的路径是连续的、平滑的．避障的，不会陷入U形障碍物，与障碍物形状和所处位置无关，能对快速变化的环境做出迅速反应。但在当前位置邻近位置中具有最大活性值的位置不惟一的情况下，产生路径可能不理想，即到达目标点的避障路径是较长的，而不是最短或者是接近最短的。文中对该不足进行了分析，并提出了改进方法，使生成路径是最短的或是接近最短。对改进方法进行了仿真，实验结果证明该方法是有效的和可行的。     

基于并联腿机构的四足仿生机器人开发

本论文主要对生物的节律运动控制机理进行仿生,模仿生物中枢模式发生器(CPG)产生节律运动控制信号,并对其工程模型进行了简化,方便控制方法实现。同时在机器人腿部结构方面,采用了具有高精度和高刚度的并联闭环机构,提高了机器人的承载能力。经实验表明,基于此并联机构腿的四足机器人可以灵活的实现平面、坡面的多种步态的稳定快速运动。     

基于GPS的仿生六足机器人实时导航定位

根据微小型仿生六足机器人的作业任务和工作环境要求,搭建一套基于嵌入式数字信号处理器TMS320VC 5509A DSP和Superstar II GPS接收板卡的机器人导航系统。采用Marconi Binary串行数据传输标准协议采集全球定位系统(GPS)的定位数据,提出用切面投影定位法对GPS提供的定位数据进行坐标转换,并根据转换后的结果对仿生六足机器人的航迹进行修正。实验结果表明该导航定位系统具有良好的实时性和定位精度。     

Inventing a Biologically Inspired, Energy Efficient Micro Aerial Vehicle

In recent years, research efforts have focused on the design, development and deployment of unmanned systems for a variety of applications ranging from intelligence and surveillance to border patrol, rescue operations, etc. Micro Aerial Vehicles are viewed as potential targets that can provide agility and accurate small area coverage while being cost-effective and can be easily launched by a single operator. The small size of MAVs allows such flight operations within confined space but the control effectors must provide sufficient maneuverability, while maintaining stability, with only limited sensing capability onboard the platform. To meet these challenges, researchers have long been attracted by the amazing attributes of biological systems, such as those exhibited by birds and insects. Birds can fly in dense flocks, executing rapid maneuvers with g-loads far in excess of modern fighter aircrafts, and yet never collide with each other, despite the absence of air traffic controllers. This paper introduces a novel framework for the design and control of a Micro Air Vehicle. The vehicle’s conceptual design is based on biologically-inspired principles and emulates a dragonfly (Odonata–Anisoptera). A sophisticated multi-layered Hybrid & Linear/Non-Linear controller to achieve extended flight times and improved agility compared to other Rotary and Flapping Wing MAV designs. The paper addresses the design and control features of the proposed QV design and gives an overview on the developmental efforts towards the prototyping of the flyer. The potential applications for such a high endurance vehicle are numerous, including air-deployable mass surveillance in cluster and swarm formations. The disposable nature of the vehicle would help in battle-field deployment as well, where such a MAV would be made available to soldiers for proximity sensing and threat level assessment. Other applications would include search and rescue operations and civilian law-enforcement.     

Surface-Tension-Driven Biologically Inspired Water Strider Robots: Theory and Experiments

Recent biological studies on water strider insects revealed how they maintain stability and maneuver on the surface of water. While macroscale bodies use buoyancy, these very small insects use surface tension force to balance their weight on water. This paper proposes a biologically inspired miniature robot that utilizes the unique scaling advantage of these insects. The paper focuses on understanding the physics of the interaction between the insect and the surface of water and on designing a robot that mimics their key features. Hydrophobic Teflon coated wire legs optimized to take the most advantage of the surface tension force are used to support the weight of the 1-g robot. It is shown that twelve of these legs can support up to 9.3 g of payload. A T-shape actuation mechanism with three piezoelectric unimorph actuators is designed and studied to enable controlled locomotion. Static and dynamic properties of the robot are analyzed and compared with the experimental results. The tethered robot can successfully make both linear and rotational motions. Maximum forward speed is measured to be 3 cm/s, and the rotational speed is 0.5 rad/s. This robot proposes a new way of locomotion on water surface for future robots and devices.     

Self-Configuration of Network Services with Biologically Inspired Learning and Adaptation

This paper proposes a self-organizing scheme based on ant metaheuristics to optimize the operation of multiple classes of managed elements on an Operations Support Systems (OSSs) for mobile pervasive communications. Ant metaheuristics are characterized by learning and adaptation capabilities against dynamic environment changes and uncertainties. As an important division of swarm agent intelligence, it distinguishes itself from centralized management schemes due to its features of robustness and scalability. We have successfully applied ant metaheuristics to the network service configuration process, which is simply redefined as: the managed elements represented as graphic nodes, and ants traverse by selecting nodes with the minimum cost constraints until the eligible network elements are located along near-optimal paths—the located elements are those needed for the configuration or activation of a particular product and service. Although the configuration process is non-transparent to end users, the negotiated SLAs between users and providers affect the overall process. This proposed self-organized learning and adaptation scheme using Ant Colony Optimization (ACO) is evaluated by simulation in Java. A performance comparison is also made with a class of Genetic Algorithm known as PBIL. Finally, the simulation results show the scalability and robustness capability of autonomous ant-like agents able to adapt to dynamic networks.     

The First Takeoff of a Biologically Inspired At-Scale Robotic Insect

Biology is a useful tool when applied to engineering challenges that have been solved in nature. Here, the emulous goal of creating an insect-sized, truly micro air vehicle is addressed by first exploring biological principles. These principles give insights on how to generate sufficient thrust to sustain flight for centimeter-scale vehicles. Here, it is shown how novel manufacturing paradigms enable the creation of the mechanical and aeromechanical subsystems of a microrobotic device that is capable of Diptera-like wing trajectories. The results are a unique microrobot: a 60 mg robotic insect that can produce sufficient thrust to accelerate vertically. Although still externally powered, this micromechanical device represents significant progress toward the creation of autonomous insect-sized micro air vehicles.     

基于改进生物启发神经网络的机器人目标搜索方法

针对未知环境下机器人目标搜索的问题,按照机器人能力不同对搜索区域进行划分,目标点在自己运动的过程中会在局部范围内留下信息素并且这些信息素会随着时间的流失而减少,机器人可以探测到这些信息素的多少进而影响机器人下一个搜索位置的选择。本文采用改进生物启发神经网络选取机器人探索范围内活性值最大的点作为下一个搜索位置。为了防止在连续的时间段内多次选择相同的点,引入禁忌搜索,把多次选择相同的点放入禁忌表中,可以有效防止陷入局部最优点。与随机搜索方式和原始的生物启发神经网络进行对比,验证了该方法对动态目标的搜索具有良好的效果。     

基于知识的银行票据二值化方法

本文结合银行票据OCR系统的开发,提出一种基于知识进行银行票据二值化的新思路,并针对各类识别域具体构造了一整套二值化方法。通过在银行票据OCR系统中的应用,验证了本文二值化方法的效果。     

Biologically Inspired Control Of A Fleet Of Uavs With Threat Evasion Strategy

In this paper, navigation algorithms for a fleet of multiple nonholonomic UAVs capable of evading a chasing predator and also pursuing a desired target are proposed. The proposed biologically‐inspired navigation algorithms are used to define path planning trajectories which are tracked by a designed backstepping tracking controller. We implement the group of nonholonomic UAVs in an adaptive network, specifically inspired by the relationship between a school of fish and a predator. This approach approximately simulates an air combat field. To put this in context, the aim is to use a biologically inspired algorithm along with a designed controller to achieve both target pursuance and effective evasion from a predator. This is equivalent to having multiple UAVs on the same mission of attacking a target, while also aware of a predator on pursuit. The UAVs aim to maneuver and evade the predator while also coordinating their movement and behaviors in a cooperative and coherent manner.     

Design and Development of the Lifting and Propulsion Mechanism for a Biologically Inspired Water Runner Robot

This paper describes the design and development of a novel robot, which attempts to emulate the basilisk lizard's ability to run on the surface of water. Previous studies of the lizards themselves have characterized their means of staying afloat. The design of a biomimetic robot utilizing similar principles is discussed, modeled, and prototyped. Functionally, the robot uses a pair of identical four bar mechanisms, with a 180 ^deg phase shift to achieve locomotion on the water's surface. Simulations for determining robot lift and power requirements are presented. Through simulation and experimentation, parameters are varied with the focus being a maximization of the ratio of lift to power. Four legged robots were more easily stabilized, and had a higher lift-to-power ratio than two legged robots. Decreases in characteristic length and running speed, and increases in foot diameter and foot penetration depth all cause a higher lift to power ratio. Experimental lift approached 80 gr, and experimental performance exceeded 12 gr/W for four legged robots with circular feet. This work opens the door for legged robots to become ambulatory over both land and water, and represents a first step toward robots which run on the water instead of floating or swimming.