A simulation study of vertical jumping from different starting postures.

This paper addresses the question of whether maximal vertical jump height depends on initial jumping posture. A direct dynamics computer simulation approach was used to avoid subject preference and practice effects. The human body was modeled as four rigid segments connected by ideal hinge joints, with movement constrained to the sagittal plane and driven by three single-joint torque actuators. Maximal height jumps were found for each of 125 different initial postures. For each initial posture, the optimal pattern of joint torque actuator onset times was found using a multidimensional simplex algorithm searching for maximal jump height. The model results revealed that maximal jump height is relatively insensitive to initial posture, but that the pattern of joint torque onset times necessary to effect these optimal heights varies considerably. Model kinematics indicate that the variability in onset times is necessary to allow the body to re-orient itself in different ways during the downward countermovement phase. This variable re-orientation strategy is followed by a more stereotyped upward thrust phase that is similar despite the differences in starting postures. Model center of mass, joint and segmental kinematics show many features found in experimental studies of jumping, despite the exclusive use of single torque actuators. However, a proximal-to-distal sequence of joint coordination was not found, possibly because of the omission of antagonist and bi-articular muscles. The results suggest that similar vertical jump heights should be obtained using a wide range of initial starting positions.