Patrick Lacouture

Foot contact event detection using kinematic data in cerebral palsy children and normal adults gait

Initial contact (IC) and toe off (TO) times are essential measurements in the analysis of temporal gait parameters, especially in cerebral palsy (CP) gait analysis. A new gait event detection algorithm, called the high pass algorithm (HPA) has been developed and is discussed in this paper. Kinematics of markers on the heel and metatarsal are used. Their forward components are high pass filtered, to amplify the contact discontinuities, thus the local extrema of the processed signal correspond to IC and TO. The accuracy and precision of HPA are compared with the gold standard of foot contact event detection, that is, force plate measurements. Furthermore HPA is compared with two other kinematics methods. This study has been conducted on 20 CP children and on eight normal adults. For normal subjects all the methods performed equally well. True errors in HPA (mean+/-standard deviation) were found to be 1+/-23 ms for IC and 2+/-25 ms for TO in CP children. These results were significantly (p<0.05) more accurate and precise than those obtained using the other algorithms. Moreover, in the case of pathological gaits, the other methods are not suitable for IC detection when IC is flatfoot or forefoot. In conclusion, the HPA is a simple and robust algorithm, which performs equally well for adults and actually performs better when applied to the gait of CP children. It is therefore recommended as the method of choice.

Three-dimensional joint dynamics and energy expenditure during the execution of a judo throwing technique (Morote Seoï Nage)

Abstract A new method for analysing judo throwing techniques is proposed. Beyond a solely descriptive analysis of the kinematic parameters, we determined the main active joints and segments in the execution of a judo throwing technique. This was achieved by calculating three-dimensional joint dynamics parameters. The environment of the judoka was manipulated by using a judo-specific ergometer to replace his partner. The ergometer was used together with two force sensors coupled with two force platforms, and six synchronized infrared cameras. Sixteen French athletes competing in senior national events participated in this study and executed the throwing technique Morote Seoï Nage ten times, using the Mayeur ergometer loaded with 20 kg. This load was chosen to cover variations in the athletes and represents the effective mass they have to displace during the movement. Our main aims were to evaluate the forces and moments at the main joints in three dimensions during Morote Seoï Nage, and determine the energy expenditure of the athletes during the movement. Contrary to the teaching of some judo coaches and experts, our results show that the main driving moments are generated by the lower limbs [mean 24% (s = 4) of the total moments at the knees and 29% (s = 3) at the hips] and the trunk (mean 28%, s = 3) and not the upper limbs. Moreover, our results show that most energy expenditure (mean 880 J, s = 160) occurs during the Tsukuri phase, when the Tori (the person who throws) is positioned under the Uke (the person who is thrown).

From human motion capture to humanoid locomotion imitation application to the robots hrp-2 and hoap-3

This paper presents a method to generate humanoid gaits from a human locomotion pattern recorded by a motion capture system. Thirty seven reflective markers were fixed on the human subject skin in order to get the subject whole body motion. To reproduce the human gait, especially the toes and heel contacts, the front and back edges of the robots feet are used as support at the start and the end of the double support phase. The balance of the robot is respected using the zero moment point (ZMP) criterion and confirmed by the simulation software OPENHRP (General Robotics, Inc®). First, the feet trajectory as well as the ZMP reference trajectory are defined from the motion of the robot controlled as a marionette with the measured human joint angles. Then a specific inverse kinematic (IK) algorithm is proposed to find the humanoid robots joint trajectories respecting the constraints of balance, floor contacts, and joint limits. The studied motion presented in this paper is a human walking trajectory containing a start, a movement in a straight line, a stop, and a quarter turn. The method was developed to be easily used for human-like robots of different sizes, masses, and structures and has been tested on the robot HRP-2 (AIST, Kawada Industries, Inc®) and on the small-sized humanoid robot HOAP-3 (Fujitsu Automation Ltd®).

The convex wrapping algorithm: a method for identifying muscle paths using the underlying bone mesh.

Associating musculoskeletal models to motion analysis data enables the determination of the muscular lengths, lengthening rates and moment arms of the muscles during the studied movement. Therefore, those models must be anatomically personalized and able to identify realistic muscular paths. Different kinds of algorithms exist to achieve this last issue, such as the wired models and the finite elements ones. After having studied the advantages and drawbacks of each one, we present the convex wrapping algorithm. Its purpose is to identify the shortest path from the origin to the insertion of a muscle wrapping over the underlying skeleton mesh while respecting possible non-sliding constraints. After the presentation of the algorithm, the results obtained are compared to a classically used wrapping surface algorithm (obstacle set method) by measuring the length and moment arm of the semitendinosus muscle during an asymptomatic gait. The convex wrapping algorithm gives an efficient and realistic way of identifying the muscular paths with respect to the underlying bones mesh without the need to define simplified geometric forms. It also enables the identification of the centroid path of the muscles if their thickness evolution function is known. All this presents a particular interest when studying populations presenting noticeable bone deformations, such as those observed in cerebral palsy or rheumatic pathologies.

Validation of a specific machine to the strength training of judokas.

With regard to judo players, like all sport activities, strength training can be divided into general and specific strength training. The specific exercises must correspond with the competitive movement. In addition, in terms of structure, they must correspond with regard to the strength time sequence, and they may be executed with overload. With respect to these important elements, we have envisaged the use of a judo-specific machine. The purpose of this study was to validate this judo-specific machine with regard to the strength training of judokas. To that end, we have measured the maximal pulling forces applied at each hand of judokas (n = 18), playing with the judo-specific machine and with a real partner. A significant difference was found between the maximal pulling forces (F collar and F sleeve) obtained utilizing the judo- specific machine (from 4.9 ± 0.4 N·kg-1 to 6.4 ± 0.3 N·kg-1 for F collar; from 4.8 ± 0.2 N·kg-1 to 6.3 ± 0.3 N·kg-1 for F sleeve) and performing with a partner (2.7 ± 0.2 N·kg-1 for F collar; 2.5 ± 0.3 N·kg-1 for F sleeve). This can be explained by the fact that the partner opposes a low resistance during the judo-throwing technique in comparison with the judo-specific machine. These results show that the judo-specific machine might be used by the judokas to execute specific exercises with overload.

Improvement of upper extremity kinematics estimation using a subject-specific forearm model implemented in a kinematic chain

Human movement reconstruction is still difficult due to noise generated by the use of skin markers. The a priori definition of a kinematic chain associated with a global optimisation method allows reducing these deleterious effects. When dealing with the forearm, this approach can be improved by personalising the two axes of rotation because their common modelling is not representative of joint geometry. The aim of the present study is to evaluate the kinematic effects of personalising these two axes of rotation, determined by a functional method and implemented in a kinematic chain (AXIS model). The AXIS model was compared with a reference model (ISB model), in which the forearm axes of rotation were defined according to the recommendations of the International Society of Biomechanics. The kinematic comparison (15 subjects and 3 tasks) was based on marker residuals (actual versus model-determined), joint kinematic root mean square differences (AXIS versus ISB) and joint amplitudes (AXIS versus ISB). The AXIS model improved the pose of the forearm and hand. The reduction in marker residuals for these segments ranged between 23% and 60%. The use of a functional method was also beneficial in personalising the flexion-extension and pronation-supination axes of the forearm. The contribution of pronation-supination, in terms of joint amplitudes, was increased by 15% during the specific task. The approach developed in this study is all the more interesting since this forearm model could be integrated into a kinematic chain to be used with a global approach becoming increasingly popular in biomechanics.

An auto-adaptable algorithm to generate human-like locomotion for different humanoid robots based on motion capture data

The work presented in this paper deals with the generation of trajectories for humanoid robots imitating human gaits captured with a motion capture system. Once the human motion is recorded, this one is modified to be adapted to the robot morphology. The proposed method could be used for human-like robots of different sizes and masses. The generated gaits are closed to the humans ones while respecting the robot balance and the floor contacts. First the human joint angles are computed from the markers coordinates and applied directly to the robot kinematics model. Then, from this non-corrected motion, the trajectories of both feet and of the Zero Moment Point (ZMP) are generated respecting the constraints of floor contact and balance control. From this data, an inverse kinematic algorithm is used to compute the joint angles of the robot according to the feet and ZMP trajectories. The results with the robot HRP-2 (AIST, Kawada Industries, Inc) and the small-sized humanoid HOAP-3 (Fujitsu Automation Ltd) are compared with the human motion.

Comparison of calibration methods for accelerometers used in human motion analysis.

In the fields of medicine and biomechanics, MEMS accelerometers are increasingly used to perform activity recognition by directly measuring acceleration; to calculate speed and position by numerical integration of the signal; or to estimate the orientation of body parts in combination with gyroscopes. For some of these applications, a highly accurate estimation of the acceleration is required. Many authors suggest improving result accuracy by updating sensor calibration parameters. Yet navigating the vast array of published calibration methods can be confusing. In this context, this paper reviews and evaluates the main measurement models and calibration methods. It also gives useful recommendations for better selection of a calibration process with regard to a specific application, which boils down to a compromise between accuracy, required installation, algorithm complexity, and time.

A generalized 3D inverted pendulum model to represent human normal walking

This paper compares different inverted pendulum models to represent the stance phase of human normal walking. We have developed a model which takes into account the mechanism of the foot during the single support phase, by defining a pivot point under the ground level. Similarly to other models, the pivot point as well as the rod length remain constant during the complete single support phase. Lowering the position of the pivot point allows reducing the vertical amplitude of the center of mass (CoM) trajectory and therefore approaching the real CoM trajectory. We have measured the whole body kinematics of a representative healthy male subject and set a reference CoM trajectory based on multi-body modeling of the human body (16 segments). Then, we have determined a common mathematical definition of two inverted pendulum models extended in the three dimensional space: the classical IP-3D and our GIP-3D model.

Two-dimensional kinematic and dynamic analysis of a karate straight punch.

The mechanical effect of punches performed in martial arts or boxing sports has been studied on different ways: the impact force was either directly measured with sensors fixed on rigid frames [5, 6] or indirectly estimated from the mechanical features of materials broken during strike tests [2, 3]. A few authors developed experimental devices for measuring this force in actual fighting conditions [1] or on a punching bag [4]. Among the studies that analyzed the kinematics of the striking segments [2, 3, 5], only one [2] related kinematic and dynamic data through the linear momentum. According to this approach, a straight punch struck by a karateka (1.68 m, 68 kg, 3rd dan black belt) on a training instrument traditionally used in karate (makiwara) was analyzed in two dimensions. The peak force was two to three times lower than the maximum values (4000 to 6000 N) reported in previous studies [1, 5, 6], which could be explained by the makiwara flexibility. The large difference between the variation of the karatekas linear momentum and the linear impulse of the target-block pointed out the limitation of a 2-D analysis of this movement, which cannot take into account the angular momenta of the trunk and the upper limbs around the vertical axis.