A side-to-side difference in the muscle size of the rectus abdominis has been suggested to increase the strain injury risk. Attenuating the difference in size of the rectus abdominis may decrease the injury risk. To explore ways to highly activate one side of the rectus abdominis, we aimed to clarify the activity levels of both sides of the muscle during asymmetric abdominal exercises. Fifteen male sprinters performed the following five asymmetric exercises for the right and left sides: (i) sit-up twist, (ii) oblique leg raise, (iii) side bridge, (iv) side bridge roll out with the elbow, and (v) side bridge roll out with the foot. Side bridge roll out with the elbow and that with the foot were performed using a wheeled platform. During the exercises, electromyographic signals were recorded bilaterally from the upper, central, and lower portions of the rectus abdominis. We calculated the root mean square of electromyograms during the concentric and eccentric phases of the exercises and normalized to that during maximal voluntary contractions. In all portions of the rectus abdominis, the root mean squares of electromyograms were significantly higher in the moving side than in the non-moving side during the concentric and eccentric phases of the side bridge, the side bridge roll out with the elbow and that with the foot (all p < 0.01), but not in sit-up twist or oblique leg raise. The root mean squares of electromyograms of all portions of the rectus abdominis in the moving side were significantly higher in the side bridge roll out with the elbow and that with the foot than in the side bridge during both phases (all p < 0.01). The results suggest that the application of the wheeled platform to side bridge is useful to highly activate one side of the rectus abdominis.