An environmental adaptive control system of a wheel type mobile robot for the rough terrain movement

Sato, Masanori; Kanda, Atsushi; Ishii, Kazuki

doi:10.1109/iros.2007.4399604

Cited by 9 publications

(6 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the situation it is facing according to its means of perception, as it will be described in more detail in section IV-B. Then, let a ∈ [ [1,7]] be the variable representing the action that can be chosen among the seven possibilities defined in section III. The robot has to select the action a to execute according to its state s : this behavioral mapping is called the policy π.…”

Section: Learning Protocol a Reinforcement Learning Algorithmmentioning

confidence: 99%

“…Machine learning techniques have already been tested on wheel-on-leg robots for identifying the terrain type among very few classes in order to select whether the proper wheel speed control [7] or the predefined gait to use [8]. However, to our knowledge, the combination of compliant behavior and reinforcement learning in order to produce obstacle negotiation strategies has never been explored before.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Obstacle negotiation learning for a compliant wheel-on-leg robot

Bouton

Grand

Benamar

2017

2017 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Building versatile control for wheel-on-leg robots on various uneven terrains still constitutes a challenge. In this paper, we propose to combine a continuous state space Q-learning algorithm with a compliant wheeled locomotion structure capable of measuring forces applied by the environment. These forces are used to choose a behavior from the modulation of actuation distribution to simple configuration adjustments. Then, the robot does not need any prior knowledge on the ground geometry and is able to react to non-anticipated obstacles. The learned policy proves to be generic and allows the robot to negotiate complex obstacles that were not seen during learning.

show abstract

Section: Learning Protocol a Reinforcement Learning Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Obstacle negotiation learning for a compliant wheel-on-leg robot

Bouton

Grand

Benamar

2017

2017 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“…PMVs can traverse continuous terrain such as a gently rolling road; however, travelling over a discontinuous terrain such as those containing bumps is difficult. To overcome these problem, numerous robots have been developed for traveling over such rough terrain, for example, legged robot such as the TITAN series [4], the hexapod robot [5] and the Tekken 2 [6], wheeled robot such as the ASGUARD [7], Shrimp [8], and Zaurus [9], and legged-wheel robot such as the Work Partner [10], Roller Walker [11], and Chariot [12]. These robots can traverse rough terrain, however, their mechanism and control are complex.…”

Section: Introductionmentioning

confidence: 99%

Proposal for an IR system to support automatic control for a personal mobility vehicle

Fujikawa

Nakajima

2012

2012 IEEE International Conference on Robotics and Biomimetics (ROBIO)

View full text Add to dashboard Cite

We propose a supporting system for automatic control of personal mobility vehicles (PMVs) to pass through a narrow space in indoor environment. This system consists of IR markers located on an entrance of a narrow space and an IR sensor to mount on a PMV. An IR sensor calculates relative distance between itself and an IR marker and angle of approach to pass through on a marker. To validate the proposed system we perform both experiments to estimate accuracy of calculated relative distance and angle and to demonstrate automatic control using our manufactured PMV. These results show that the proposed IR system is effective in path guidance for PMVs.

show abstract

“…Owing to their stability, maneuverability, and simple controls, wheels are the most frequently used mechanism for exploration rovers. Examples of wheeled mobile robots are Micro5 (Kubota et al 2003), Rocky7 (Volpe et al 1997), Shrimp (Siegwart et al 2002), CRAB (Thueer et al 2006), and Zaurus (Sato et al 2007). These have passive linkage mechanisms.…”

Section: Introductionmentioning

confidence: 99%

RT-Mover: a rough terrain mobile robot with a simple leg–wheel hybrid mechanism

Nakajima¹

2011

The International Journal of Robotics Research

View full text Add to dashboard Cite

There is a strong demand in many fields for practical robots, such as a porter robot and a personal mobility robot, that can move over rough terrain while carrying a load horizontally. We have developed a robot, called RT-Mover, which shows adequate mobility performance on targeted types of rough terrain. It has four drivable wheels and two leg-like axles but only five active shafts. A strength of this robot is that it realizes both a leg mode and a wheel mode in a simple mechanism. In this paper, the mechanical design concept is discussed. With an emphasis on minimizing the number of drive shafts, a mechanism is designed for a four-wheeled mobile body that is widely used in practical locomotive machinery. Also, strategies for moving on rough terrain are proposed. The kinematics, stability, and control of RT-Mover are also described in detail. Some typical cases of rough terrain for wheel mode and leg mode are selected, and the robot's ability of locomotion is assessed through simulations and experiments. In each case, the robot is able to move over rough terrain while maintaining the horizontal orientation of its platform.

show abstract

An environmental adaptive control system of a wheel type mobile robot for the rough terrain movement

Cited by 9 publications

References 15 publications

Obstacle negotiation learning for a compliant wheel-on-leg robot

Obstacle negotiation learning for a compliant wheel-on-leg robot

Proposal for an IR system to support automatic control for a personal mobility vehicle

RT-Mover: a rough terrain mobile robot with a simple leg–wheel hybrid mechanism

Contact Info

Product

Resources

About