The global rise in demand for critical metals has renewed interest in deep-sea mining of polymetallic nodules on the abyssal plains. Commercial concepts primarily adopt vehicles with tracks to collect these nodules from the soft cohesive seabeds at depths exceeding 4,500 m. These vehicles need to generate sufficient traction and maintain mobility while minimizing seabed disturbance. Predicting their performance, however, remains difficult because of the complex soil behavior and the limited amount of experimental data available. Most existing mobility models are empirical or assume perfectly plastic soils, which neglect strain-dependent softening and remolding.

This thesis develops a generic analytical framework that connects soil stress–strain behavior to track-scale mobility. The model couples a bearing-capacity-based sinkage formulation with a saturated soft-plastic shear–displacement relation. It allows the use of peak and residual shear strengths, as well as the characteristic displacement to peak strength (Kω), directly as input parameters. The framework handles traction and resistance under quasi-static, undrained conditions and applies to both straight-line and turning motion.

The results show that soil sensitivity (the ratio between peak and residual strength) and Kω control the shape of the traction-slip curve. Larger Kω shifts the peak to higher slip and delays remolding, while higher sensitivity steepens the post-peak drop and lowers available traction once the soil is remolded. Increasing the effective contact length improves traction up to a plateau, after which the additional gain becomes small. In heterogeneous (layered) soils, where strength increases with depth, traction increases with grounding pressure but at a decreasing rate. In homogeneous (uniform) soils, traction decreases with added normal load because increased grounding pressure limits shear mobilization before failure.

During turning, most of the track footprint exceeds the characteristic displacement Kω and the response becomes governed by residual strength. Turning is therefore traction-limited and controlled mainly by soil sensitivity, residual strength and geometry. A clear transition in the minimum turning radius is observed when the required thrust exceeds the residual traction level; beyond this point, small-radius turns become infeasible.

The results indicate that mobility can only be predicted reliably when post-peak soil behavior is included. Turning design should be based on remolded soil conditions, while straight-line operation should remain in the pre-peak region to prevent stalling. Design efforts should focus on optimizing contact length, adjusting grouser height to local seabed conditions, and measuring residual strength and sensitivity for the specific site.