**3. Results**

#### *3.1. Temporal Stride Kinematics*

As the load mass and grade increased, subjects took a longer step length than level walking (Table 1). The changes in the normalized step widths were different between the load mass and grade groups. Both of the temporal stride kinematics were not significantly different from level walking without backpack loads (all *p* > 0.05).

**Table 1.** Normalized step lengths and step widths during different grades of walking with different load masses.


All the step lengths and step widths were divided by the values during level walking without backpack loads. Therefore, the normalized step lengths and step widths were "1.00", and all the normalized step lengths and step widths were dimensionless in Table 1.

#### *3.2. Muscle Activities*

As expected, the mean muscle activities of the hip, knee, and ankle extensors generally increased with the increase of the load mass and grade (Figures 2–5 and Table S1). Both the grade and load mass had a significant effect on all muscles (*p* < 0.05), and none of the muscles showed significant grade-load-mass interactions (*p* > 0.05).

**Figure 2.** The mean EMG signals for muscles during different slope walking across all load masses in one gait from right heel strike to the next right heel strike, normalized to the mean activity during level walking without backpack loads. \* Representing the mean EMG was significantly different from level walking across all the backpack loads, according to post hoc comparisons with a Bonferroni adjusted level of significance (*p* < 0.0072). The red line represented the normalized mean EMG of each muscle at different grades without backpack loads. The yellow one represented the normalized mean EMG of each muscle at different grades with 10-kg backpack loads. The green one represented the normalized mean EMG of each muscle at different grades with 20-kg backpack loads. The blue one represented the normalized mean EMG of each muscle at different grades with 30-kg backpack loads.

**Figure 3.** The muscle activity of leg muscles in one gait during different slope walking across all backpack loads, normalized to the mean activity during level walking without backpack loads: (**a**) The muscle activity of the hip extensor muscles. (**b**) The muscle activity of the knee extensor muscles. (**c**) The muscle activity of the ankle muscles. The different colors of the curves had the same representation as those colors in Figure 2. The gray area represented the higher muscle activation duration in one gait. RMS: root mean square.

**Figure 4.** Mean EMG signals for muscles while walking across all grades with different backpack loads during one gait from right heel strike to the next right heel strike, normalized to the mean activity during level walking without backpack loads. \* Representing the mean EMG was significantly different from walking without backpack loads across all the grades, according to post hoc comparisons with a Bonferroni adjusted level of significance (*p* < 0.0072). The line in dark gray represented the normalized mean EMG of each muscle with different load masses during level walking. The line in orange represented the normalized mean EMG of each muscle with different load masses during slope walking at grade 3◦. The line in purple represented the normalized mean EMG of each muscle with different load masses during slope walking at grade 5◦. The line in blue represented the normalized mean EMG of each muscle with different load masses during slope walking at grade 10◦.

**Figure 5.** The muscle activity of leg muscles in one gait during walking with different backpack load masses across all grades, normalized to the mean activity during level walking without backpack loads: (**a**) The muscle activity of the hip extensor muscles. (**b**) The muscle activity of the knee extensor muscle. (**c**) The muscle activity of the ankle muscles. The different colors of the curves had the same representation as those colors in Figure 4. The gray area represented the higher muscle activation duration in one gait.

#### 3.2.1. Grade E ffects

The mean EMG signals of the hip, knee, and ankle extensor muscles increased generally during most of the slope walking, especially during the 10◦ slope walking (GM, *p* = 1.32 × <sup>10</sup>−8; HA, *p* = 2.33 × <sup>10</sup>−16; RF, *p* = 0.2 × <sup>10</sup>−5; and GA, *p* = 6.74 × <sup>10</sup>−9). Compared to the level walking, these increases were statistically significant for GM, RF, and AT at the 10◦ grade and for HA and GA at the 5◦ and 10◦ grades (Figure 2). Compared to the level walking with the same loads, the mean EMG of GM increased greatly during walking at the 10◦ grade by 46% to 207%, HA increased by 110% to 226%, RF increased by 44% to 203%, AT increased by 48% to 68%, and GA increased by 30% to 100% (shown in Table S2).

As the grade increased, the muscle activity of the hip extensor muscles (GM and HA, especially GM) increased both in the activation value and duration (Figure 3a). The maximum activation value of the GM changed from 0.52 to 1.24, and the activation duration increased from about 28% to 48% of one gait as the grade increased. The maximum activation value of the HA changed from 0.52 to 0.81, and the activation duration changed from 20% to about 45% of one gait as the grade increased.

The muscle activity of the knee extensor muscle (RF) increased most in the activation value, shown in Figure 3b. The maximum activation value of the RF increased from 0.82 to 1.60 during the early stance stage and from 0.4 to about 0.9 during the early swing stage.

As expected, the muscle activity of the ankle extensor (GA) increased considerably in the muscle activation value (Figure 3c). The maximum activation of the GA during the median and the late stances increased highly, from 0.52 to 1.07. Compared to the ankle extensor, the muscle activity of the ankle dorsiflexion (AT) increased slightly, from 0.48 to 0.57 during the early stance stage and swing stage.

#### 3.2.2. Load Mass E ffects

The mean EMG signals of the hip, knee, and ankle extensor muscles increased generally as the backpack load masses increased. Compared to walking without backpack loads, the increases were statistically significant for the GM and GA during walking with 30-kg loads (GM, *p* = 0.000091 and GA, *p* = 0.00091) and for the RF during walking with 30 kg (*p* = 8.86 × <sup>10</sup>−7) (Figure 4). The increases in the HA and AT were not statistically significant (*p* > 0.0072). Compared to walking without loads at the same grade, the mean EMG of the GM increased by 5% to 173% with a 30-kg backpack load, the HA increased by −5% to 26%, the RF increased by 104% to 172%, the AT increased by 0% to 35%, and the GA increased by 15% to 61% (the data are shown in Table S3).

The muscle activity of the hip extensors (GM and HA) increased as the backpack load mass increased, as shown in Figure 5a. The maximum muscle activity of the GM increased greatly from 0.66 to 1.24 as the backpack load mass increased from 0 to 30 kg. The maximum muscle activity of the HA increased slightly from 0.63 to 0.81 as the load mass increased.

The knee extensor muscle RF increased greatly in muscle activation as the backpack load mass increased (Figure 5b). The maximum muscle activation of the RF increased greatly from 0.67 to 1.60 during the early stance stage as the load mass increased from 0 to 30 kg across all the slope walking.

The ankle extensor muscle GA increased greatly in muscle activation from 0.70 to 1.07 as the backpack load mass increased from 0 to 30 kg, and the dorsiflexion muscle AT increased slightly, from 0.50 to 0.58 (Figure 5c). The muscle activity demonstrated that the ankle extensor muscles were activated more than the flexor muscles during slope walking with backpack loads.
