Sensor-driven neural control for omnidirectional locomotion and versatile reactive behaviors of walking machines

Manoonpong, Poramate; Pasemann, Frank; Wörgötter, Florentin

doi:10.1016/j.robot.2007.07.004

Cited by 68 publications

(98 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This learned behavior together with the escape response is controlled by adaptive sensor-driven neural control using a correlation based differential Hebbian learning rule. The work presented here extends our previous works [3], [4] by integrating this learning mechanism into the original modular neural control [3], [4] leading to the adaptive behavior. However, the main purpose of this article is not only to demonstrate the biologicallyinspired learning on the walking machine (adaptive behavior) but also to show that the adaptive sensor-driven neural control can be a powerful technique to solve sensorimotor coordination problems of many degrees-of-freedom systems and to effectively provide an online learning capability to the systems.…”

Section: Introductionsupporting

confidence: 59%

Adaptive Sensor-Driven Neural Control for Learning in Walking Machines

Manoonpong

Wörgötter

2009

Neural Information Processing

View full text Add to dashboard Cite

Abstract. Wild rodents learn the danger-predicting meaning of predator bird calls through the paring of cues which are an aversive stimulus (immediate danger signal or unconditioned stimulus, US) and the acoustic stimulus (predator signal or conditioned stimulus, CS). This learning is a form of pavlovian conditioning. In analogy, in this article a setup is described where adaptive sensor-driven neural control is used to simulate biologically-inspired acoustic predator-recognition learning for a safe escape on a six-legged walking machine. As a result, the controller allows the walking machine to learn the association of a predictive acoustic signal (predator signal, CS) and a reflex infrared signal (immediate danger signal, US). Such that after learning the machine performs fast walking behavior when "hearing" an approaching predator from behind leading to safely escape from the attack.

show abstract

Section: Introductionsupporting

confidence: 59%

Adaptive Sensor-Driven Neural Control for Learning in Walking Machines

Manoonpong

Wörgötter

2009

Neural Information Processing

View full text Add to dashboard Cite

show abstract

“…The proposed system has been applied to two simulated robots: the wheeled robot NIMM [2] and the hexapod robot AMOS [5].…”

Section: Resultsmentioning

confidence: 99%

“…The proposed system is used to control the simulated hexapod robot AMOS [5]. AMOS's walking ability relies on the movements of its six legs which are controlled by signals received at the legs' joints and produced by a two neurons oscillator (CPG) [5].…”

Section: Experiments 3: Amos In a Multiple-goal Environmentmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Landmark-Based Navigation System Using Learning Techniques

Zeidan

Dasgupta

Wörgötter

et al. 2014

From Animals to Animats 13

View full text Add to dashboard Cite

Abstract. The goal-directed navigational ability of animals is an essential prerequisite for them to survive. They can learn to navigate to a distal goal in a complex environment. During this long-distance navigation, they exploit environmental features, like landmarks, to guide them towards their goal. Inspired by this, we develop an adaptive landmarkbased navigation system based on sequential reinforcement learning. In addition, correlation-based learning is also integrated into the system to improve learning performance. The proposed system has been applied to simulated simple wheeled and more complex hexapod robots. As a result, it allows the robots to successfully learn to navigate to distal goals in complex environments.

show abstract

“…This category consists of robotic platforms that uses elaborate sensing elements to estimate the terrain type and use adaptive algorithms to plan its route, gait and speed [6]. RiSE [7] robot is one of the example for this category.…”

Section: A Closed Loop Gait Planning Designsmentioning

confidence: 99%

IONS: A Quadruped Robot for Multi-terrain Applications

Agrawal¹,

Dagale²,

Mohan³

et al. 2015

IJMMM

View full text Add to dashboard Cite

Abstract-In this paper we present the design and development of the IONS (Indoor Outdoor Navigation and Surveillance), a versatile, low-cost, power autonomous, statically stable quadruped robot. The robot has unique compliant wheeled-leg limb design that enables it to climb up and descend stairs of various riser heights. It is designed for statically stable walking, turning, climbing and descending gaits using minimal terrain sensing and simple clock driven open-loop control mechanism. It is equally effective in indoor as well as outdoor settings. Its salient features include auto-alignment of the body while climbing stairs, distinction between stairs and obstacles, collision avoidance etc. The robot is controlled using a web based user interface under ubiquitous Wi-Fi network. The operator use live video feed streamed from on-board camera mounted on the robot platform for gait regulation. On-board system keeps track of signal strength, detects communication failure and sends health status report of the robot to the operator console at regular intervals.

show abstract

Sensor-driven neural control for omnidirectional locomotion and versatile reactive behaviors of walking machines

Cited by 68 publications

References 44 publications

Adaptive Sensor-Driven Neural Control for Learning in Walking Machines

Adaptive Sensor-Driven Neural Control for Learning in Walking Machines

Adaptive Landmark-Based Navigation System Using Learning Techniques

IONS: A Quadruped Robot for Multi-terrain Applications

Contact Info

Product

Resources

About