Fig 2 - uploaded by Wael Suleiman
Content may be subject to copyright.
Source publication
Context in source publication
Similar publications
Abstract- Facial emotion recognition has an important position in the computer vision and artificial intelligence field. In addition, real-time face recognition applications have to be able to be performed at high speed and accuracy rate in order to make human-computer interaction successful in increasing artificial intelligence and humanoid robot...
En este artículo se presenta una clasificación de los generadores de patrones para el caminado de robots humanoides más utilizados en la literatura. El trabajo inicia con una breve descripción de los conceptos básicos del caminado de los robots humanoides a fin de familiarizar al lector con el tema. Posteriormente se presenta la clasificación de lo...
Citations
... Piperakis et al. (Piperakis & Trahanias, 2016) used nonlinear dynamics of CoM as process model to estimate three-dimensional external force based on extended Kalman filter (EKF), and improved the accuracy by considering the influence of the rate of angular momentum in Piperakis et al. (2018). To simplify the observer and make it easy to run in real-time with the embedded processor of low-cost biped robots, Hawley et al. (Hawley & Suleiman, 2016) explored the relationship between external force and ZMP, and fused the measurements from IMU and FSRs to estimate the external force. The model of the observer is linear, thus the filter is implemented by simple KF. ...
Push recovery is of vital importance for biped robots due to the complexity of the non-experimental environment. This paper proposes an external force observer aided push recovery control framework for the inexpensive biped robot. Initially, a linear external force observer is designed. This observer used centroid dynamics as filter model, thus the effect of the rate of angular momentum is considered. The linear process model guarantees the observer is easy to run in real-time with low computational resources. The force/torque (F/T) sensor which is difficult to integrate on small or inexpensive robots due to the size and cost is not necessary. Moreover, the estimated external force is fed back to the whole-body torque controller, and it is compensated by the foot contact force. Compared to the pure error-based controller, the response speed of the robot to external force is improved. This framework is verified on the defensor biped robot model developed by our team and compared to the existing approaches. The accuracy of the external force estimation and the omnidirectional push recovery ability of the biped robot are improved.
... Many researchers have designed planning and control algorithms for humanoid robots that incorporate foot sensory feedback in dynamic walking [2], self-balancing [3], push-and-recovery [4], etc. In addition, foot force sensors are also applied to estimate the humanoid robot's center of mass (CoM) location [5], and external forces [6]. Recent studies use foot force sensors to identify the mass properties of heavy objects in manipulation tasks of humanoid robots [7], [8]. ...
... ∆x also includes a first-order term of the force outputs to reduce the deviation at the same calibration spot for different applied weights, which has already demonstrated its effectiveness in our previous work [7]. The formulation is the same for ∆y in (5). Eventually, a nonlinear least squares is used to minimize the error between the corrected CoP and the ground truth: ...
... where c L , c R , n L and n R are the measured CoP and GRF using sensor parameters solved by (19) for the left and right feet. ∆s L and ∆s R are the correction terms applied to the left and right feet defined in (5). Then we use NLS to minimize the error between corrected and modeled CoPs over the parameters of the correction terms: ...
This letter presents a novel method for smaller-sized humanoid robots to self-calibrate their foot force sensors. This method consists of two steps: 1. The robot is commanded to move along several planned whole-body trajectories. 2. A nonlinear least-squares technique is implemented to determine sensor parameters by minimizing the error between the measured and modeled center of pressure (CoP) and ground reaction force (GRF) in the robot's movement. This is the first proposed autonomous calibration method for foot force sensors in humanoid robots. Additionally, a manual calibration method is developed to improve the CoP measurement accuracy, which establishes the ground truth for evaluating the self-calibration approach. The manual calibration and the self-calibration are implemented on our previously presented force-sensing shoes. The results show that the self-calibration method, without any manual intervention, can accurately estimate CoP and GRF.
... Many researchers and developers design planning and control algorithms for humanoid robots according to their foot sensory feedback in the studies of dynamic walking [2], self-balancing [3], push-and-recovery [4], etc. Additionally, many state estimation problems such as estimating center of mass (CoM) location [5] and external forces [6] in the dynamic motion of the humanoid robot require the integration of foot force data together with other in-body sensory data like an encoder or an IMU. Recent studies also show that high-precision foot force sensors can be used to identify the inertial properties of heavy objects for humanoid robot manipulation tasks [7]. ...
To function autonomously in the physical world, humanoid robots need high-fidelity sensing systems, especially for forces that cannot be easily modeled. Modeling forces in robot feet is particularly challenging due to static indeterminacy, thereby requiring direct sensing. Unfortunately, resolving forces in the feet of some smaller-sized humanoids is limited both by the quality of sensors and the current algorithms used to interpret the data. This paper presents light-weight, low-cost and open-source force-sensing shoes to improve force measurement for popular smaller-sized humanoid robots, and a method for calibrating the shoes. The shoes measure center of pressure (CoP) and normal ground reaction force (GRF). The calibration method enables each individual shoe to reach high measurement precision by applying known forces at different locations of the shoe and using a regularized least squares optimization to interpret sensor outputs. A NAO robot is used as our experimental platform. Experiments are conducted to compare the measurement performance between the shoes and the robot's factory-installed force-sensing resistors (FSRs), and to evaluate the calibration method over these two sensing modules. Experimental results show that the shoes significantly improve CoP and GRF measurement precision compared to the robot's built-in FSRs. Moreover, the developed calibration method improves the measurement performance for both our shoes and the built-in FSRs.
... Another model-based method to estimate the magnitude of an external force applied to a humanoid robot is presented in [16]. Its main advantage is that it does not require using F/T sensors but instead it uses measurements from the robot's force-sensing resistors (FSR) sensors. ...
... where 2 ... z c and 2F z are respectively the covariance of ... z c andF z . By applying Kalman filter on (16), an estimation of the state vector is obtained: ...
... CoM Acceleration ZMP CoM position Ground reaction force Kalman filter applied on Eq (16) z c (kT ) ...
... Another model-based method to estimate the magnitude of an external force applied to a humanoid robot is presented in [16]. Its main advantage is that it does not require using F/T sensors but instead it uses measurements from the robot's force-sensing resistors (FSR) sensors. ...
... where 2 ... z c and 2F z are respectively the covariance of ... z c andF z . By applying Kalman filter on (16), an estimation of the state vector is obtained: ...
... CoM Acceleration ZMP CoM position Ground reaction force Kalman filter applied on Eq (16) z c (kT ) ...
External force observer for humanoid robots has been widely studied in the literature. However, most of the proposed approaches generally rely on information from six-axis force/torque sensors, which the small or medium-sized humanoid robots usually do not have. As a result, those approaches cannot be applied to this category of humanoid robots, which is widely used nowadays in education or research.
In this paper, we propose a Kalman filter-based observer to estimate the three components of an external force applied in any direction and at an arbitrary point of the robot’s structure. The observer is simple to implement and can easily run in real-time using the embedded processor of a small or medium-sized humanoid robot, such as Nao or Darwin-OP. Moreover, the observer does not require any changes to the robot’s hardware, as it only uses measurements from the available force-sensing resistors (FSR) inserted under the feet of the humanoid robot and from the robot’s inertial measurement unit (IMU).
The proposed observer was extensively validated on a Nao humanoid robot in both cases of standing still or walking while an external force was applied to the robot. In the conducted experiments, the observer successfully estimated the external force within a reasonable margin of error. Moreover, the experimental data and the MATLAB and C++/ROS implementations of the proposed observer are available as an open-source package.
... Another model-based method to estimate the magnitude of an external force applied to a humanoid robot is presented in [16]. Its main advantage is that it does not require using F/T sensors but instead it uses measurements from the robot's force-sensing resistors (FSR) sensors. ...
... where σ 2 ... z c and σ 2F z are respectively the covariance of ... z c andF z . By applying Kalman filter on (16), an estimation of the state vector is obtained: ...
... CoM Acceleration ZMP CoM position Ground reaction force Kalman filter applied on Eq (16) z c (kT ) ...
External force observer for humanoid robots has been widely studied in the literature. However, most of the proposed approaches generally rely on information from six-axis force/torque sensors, which the small or medium-sized humanoid robots usually do not have. As a result, those approaches cannot be applied to this category of humanoid robots, which is widely used nowadays in education or research. In this paper, we propose a Kalman filter-based observer to estimate the three components of an external force applied in any direction and at an arbitrary point of the robot's structure. The observer is simple to implement and can easily run in real time using the embedded processor of a small or medium-sized humanoid robot, such as Nao or Darwin-OP. Moreover, the observer does not require any changes to the robot's hardware, as it only uses measurements from the available force-sensing resistors (FSR) inserted under the feet of the humanoid robot and from the robot's inertial measurement unit (IMU). The proposed observer was extensively validated on a Nao humanoid robot in both cases of standing still or walking while an external force was applied to the robot. In the conducted experiments, the observer successfully estimated the external force within a reasonable margin of error. Moreover, the experimental data and the MATLAB and C++/ROS implementations of the proposed observer are available as an open source package. a
... In most cases, the external force F c is obtained with six-axis F/T sensors located in the robot wrists. Even though this approach has been proven to be efficient, a lot of humanoid robots are not equipped with such sensors, thus the external force must be estimated using less reliable methods [4,14,24]. In the next section, the dynamic of the cooperative transportation task by two humanoids is analyzed. ...
This paper aims at proposing a comprehensive control framework designed for cooperative transportation of a heavy load by two humanoid robots. First, a simplified dynamic model of the cooperative task is developed and the system stability is rigorously analyzed. Next, a centralized controller is formulated, this formulation provides an optimal control of the system by considering the robots dynamical stability while satisfying the robot-object-robot constraints. Finally, the controller is integrated with an arm controller and a local planner mod- ule forming a complete framework for cooperative transportation tasks. The approach is thoroughly analyzed and validated in simulation, and experiments are carried out on a team of two Nao humanoid robots transporting a range of objects placed on a small table. The experimental results pointed out the robustness of the approach as the robots successfully accomplished the trans- portation tasks in a stable way, moreover the transported objects masses were up to half the mass of one of the robot. Besides increasing the robot payload, some of the transported objects are relatively large and it is simply impossible for a single robot to transport them.
... The framework generates a feasible trajectory for the humanoid robot and the cart while considering the physical limitations of the robot. In the proposed implementation, the robot is only able to differentiate between light and heavy loads by executing a turning in place motion with the cart, and it 1. An external force applied to a Nao robot then uses the appropriate set of motion primitives in the planning phase. ...
... Another model-based method to estimate the magnitude of an external force applied to a humanoid robot is presented in [1]. Its main advantage is that it does not require using F/T sensors but instead it uses measurements from the robot's force-sensing resistors (FSR) sensors. ...
... where the operators × and (.|.) design respectively the cross and scalar products, and 1 As a general rule, bold parameters are vectors or matrices ...
External force observer for humanoid robots has been widely studied in the literature. However, most of the proposed approaches generally rely on information from six-axis force/torque sensors, which the small or medium-sized humanoid robots usually do not have. As a result, those approaches cannot be applied to this category of humanoid robots, which is widely used nowadays in education or research. In this paper, we improve the external force observer in [1] to handle the case of an external force applied in any direction and at an arbitrary point of the robot structure. The new observer is based on Kalman filter formulation and it allows the estimation of the three force components. The observer is simple to implement and can easily run in real time using the embedded processor of a medium-sized humanoid robot such as Nao or Darwin-OP. Moreover, the observer does not require any change to the robot hardware as it only uses measurements from the available force-sensing resistors (FSR) inserted under the feet of the humanoid robot and from the robot inertial measurement unit (IMU). The proposed observer was extensively validated on a Nao humanoid robot. In all conducted experiments, the observer successfully estimated the external force within a reasonable margin of error. C++/ROS implementation and the MATLAB code that has been used to produce the results in this paper can be downloaded https://goo.gl/VkhejY . Video of the experiments is also available https://youtu.be/QxPy62EaepU
... Moreover, the friction on the wheels and the transported mass have not been taken into account in the current study. An ongoing work focuses on the integration of two techniques [10,11] that we have recently developed into the framework in order to consider the friction and transported load, we expect that the new pattern generator will allow us to increase the transported load while ensuring the robot equilibrium. ...
Although in recent years there have been quite a few studies aimed at the navigation of robots in cluttered environments, few of these have addressed the problem of robots navigating while moving a large or heavy objects. This is especially useful when transporting loads with variable weights and shapes without having to change the robot hardware. Inspired by the wide use of makeshift carts by humans, we tackle, in this work, the problem of a humanoid robot navigating in a cluttered environment while displacing a heavy load that lies on a cart-like object. We present a complete navigation scheme, from the incremental construction of a map of the environment and the computation of collision-free trajectories to the control to execute these trajectories. Our contributions are as follows: (1) a whole-body control scheme that makes the humanoid use its hands and arms to control the motions of the cart–load system (e.g. tight turns) (2) a sensorless approach to automatically select the appropriate primitive set according to the load weight (3) a motion planning algorithm to find an obstacle-free trajectory using the appropriate primitive set and the constructed map of the environment as input (4) an efficient filtering technique to remove the cart from the field of view of the robot while improving the general performances of the SLAM algorithms and (5) a continuous and consistent odometry data formed by fusing the visual and the robot odometry information. We present experiments conducted on a real Nao robot, equipped with an RGB-D sensor mounted on its head, pushing a cart with different loads. Our experiments show that the payload can be significantly increased without changing the robot's main hardware, and therefore enacting the capacity of humanoid robots in real-life situations.
... In a previous work [7], we proposed a method to estimate an external force applied in the horizontal plane. This method is mainly designed for small humanoid robots, such as Nao robot, and only uses the information from Force Sensitive Resistors (FSR) and Inertial Measurement Unit (IMU). ...
... In order to estimate the parameters of the cart, three experiments were performed with a Nao robot pushing different masses using a rudimentary wheeled-table. In each experiment, we estimated the external force applied on the robot using our force observer [7], and tried to correctly guess the mass transported on the cart by using Eq. (3) and varying the cart-parameter µ. ...
... • A mass of 5kg is then added on the cart at T = T 0 while the robot continues to walk as usual. • The robot estimates the mean of detected external forces over an interval of 4 steps (about 3 seconds) [7]. • At T = T 0 + 5s, the robot controller is triggered manually and the transported mass is estimated using Eq.(3), and then integrated into the walk pattern generator using the control law presented in Section III. ...
