Published papers

This paper proposes an energy-efficient trajectory planning method for a wheeled robot in slope ascending scenario. This method basically exploits a power consumption model of the robot which is obtained from the following two phases: first, the values of the robot power consumption in different slope angle, robot velocity, and robot heading angle are numerically calculated using a dynamic simulation of the robot taking into account of an accurate wheel-soil interaction mechanics; and subsequently, an approximated model for the robot power consumption is elaborated by neural network. Using the power consumption model an energy-efficient trajectory for a slope traversal case can be generated. Here, a bezier curve is used to provide a smooth trajectory designed with an appropriate arrival time as well as the control points for the curve. The numerical simulation result of the trajectory planning based on the proposed method gives several insights for a feasible trajectory for slope ascending scenario.

Tracked vehicles are commonly used on construction sites because of their high vehicle stability and traversability with large ground contact area. Grousers attached to the track shoe enhance vehicle stability and traversability, mitigating the risk of slipping on rough terrain. While many past studies investigated the mechanical effect of a single grouser shoe, experimental test of an entire grousered-track unit remains limited due to non-uniform stress distribution of the track unit on the ground. In this study, we developed an experimental apparatus with a grousered-track unit and conducted drawbar pull experiment on sandy terrain. We measured the slip ratio and sinkage of the track unit with various traction load corresponding to the drawbar pull of the track unit. The relationship between the drawbar pull coefficient, which is the ratio of the drawbar pull to the vertical load of the track, and the slip ratio was examined. The experiments confirmed that the slip ratio remains constant when the drawbar pull coefficient is approximately 0.4 or lower. Within this range, the track unit maintains small sinkage and a stable driving state. However, when the drawbar pull coefficient exceeds approximately 0.45, the slip ratio sharply increases as larger traction load cases the track unit to an unstable driving state by repeating the move-and-stop motion. This trend was observed regardless of the varied vertical load. These findings suggest that the relationship of the slip-drawbar pull coefficient and the threshold for stable/unstable driving state are useful for evaluating the traversability of the grousered-track unit.

Reactive mobile robot navigation in unstructured environments is challenging when robots encounter unexpected obstacles that invalidate previously planned trajectories. Model predictive path integral control (MPPI) enables reactive planning, but still suffers from limited prediction horizons that lead to local minima traps near obstacles. Current solutions rely on heuristic cost design or scenario-specific pre-training, which often limits their adaptability to new environments. We introduce dynamic repulsive potential augmented MPPI (DRPA-MPPI), which dynamically detects potential entrapments on the predicted trajectories. Upon detecting local minima, DRPA-MPPI automatically switches between standard goal-oriented optimization and a modified cost function that generates repulsive forces away from local minima. Comprehensive testing in simulated obstacle-rich environments confirms DRPA-MPPI's superior navigation performance and safety compared to conventional methods with less computational burden.

Predicting wheel mobility in low-gravity environments through Earth-based gravity tests offers a practical alternative to expensive parabolic flights and computationally intensive numerical simulations. However, an optimal scaling method for varying wheel speeds remains unidentified. This study systematically evaluated three scaling methods — Granular Scaling Laws (GSL), reduced-weight tests, and equal-mass tests — using Discrete Element Method simulations at three wheel angular velocities (π/10, π, and 2π rad/s). The methods were assessed based on their accuracy in predicting horizontal velocity, slip ratio, sinkage, and power consumption under free-driving conditions. GSL maintained errors below 5% across all conditions, while the equal-mass test showed velocity-dependent degradation with errors reaching 234% at high speeds. The reduced-weight test underestimated sinkage by over 100%, risking vehicle immobilization. An analytical framework employing an inertial number was developed to quantify soil flow characteristics, facilitating a comprehensive comparative analysis of the scaling methods. This analysis revealed that the equal-mass test inadequately captured dynamic flow phenomena, accounting for its velocity-dependent degradation. Conversely, GSL accurately reproduced soil flow characteristics under all conditions, enabling precise mobility predictions over a broad velocity range. These findings establish GSL as the most accurate and practical scaling approach for extraterrestrial rover mobility design and analysis.

Landing dynamics play a crucial role in the success of planetary missions, particularly in low-gravity environments, and determining optimal landing sites continues to be challenging. This paper reports a method for landing safety evaluation based on spacecraft dynamics for landings on the surface layer of small bodies in the Martian moons exploration project. Unlike conventional short-duration contact explorations, spacecraft designed for the exploration of Martian moons will remain on the surface of Phobos for several hours post landing. Owing to low gravitational forces, small bodies exhibit an inadequate restoring force, which can result in overturning and rebounding of the spacecraft; moreover, the gravitational forces on Martian moons are several tens of times greater than that on other small bodies, such as Ryugu and Itokawa, necessitating a consideration of the impact acceleration during landing. In this study, a distinctive model was constructed based on low-gravity experimental results, with the capacity to replicate the flexibility of the spacecraft structure, tank sloshing, shock absorption, and interaction with the surface layer. Subsequently, we confirmed the unique behaviors of the spacecraft during its landing simulation on authentic terrain, and candidate landing areas were evaluated based on nine safety criteria. This evaluation process entailed the identification of safe landing areas, yielding a 20% reduction in the number of safe landing sites compared with that based exclusively on terrain.

This paper presents the development and experimental evaluation of a bucket drum excavation system for InSitu Resource Utilization (ISRU). A custom-designed bucket drum, equipped with a vertical lift mechanism for reaction force control during excavation, has been developed and mounted on a mobile robotic platform. Experiments under varying target reaction force conditions demonstrated the controller's suitability and effectiveness. Additionally, excavation performance was compared under two scenarios: when the robot was stationary and while it was in motion. Results showed that the excavation by bucket drum in robot motion was more energy efficient since the bucket drum can excavate fresh and uncompacted soil. These findings highlight the potential of reaction force control for the bucket drum mounted on mobile robotic platforms.

This paper addresses an energy-efficient object collection problem using two mobile robots twins, equipped with a flexible tool such as a net or tether. This problem is subject to a routing problem for a collection sequence that minimizes total energy consumption of both robots for collecting scattered, multiple objects. To this end, we propose a dynamic programming-based route planning method, where the robot state is defined by energy consumption and its position. As the flexible tool allows the robots to collect a number of objects all at once, we implement a grouping circle of the tool, the radius of which depends on the number of the collected objects. The method incorporates a load-dependent velocity model as well as the load-dependent radius of object grouping circles to realistically reflect robot dynamics during the cooperative collection and transportation of the objects. Simulations were conducted under various scenarios with different object distributions, start positions of the robots, and the maximum velocities of the robots. The simulation results demonstrate that the proposed method reduces both energy consumption and travel time compared to baseline methods.

This work proposes a system capable of estimating the longitudinal velocity of a wheeled robot on loose sandy terrain using only acoustic sensors. These will likely be present anyway in many future ground mobile robots for human-robot interaction purposes. The proposed system consists of an audio feature extraction module, based on gammatone filterbanks, and a prediction module, based on a convolutional neural network. The proposed system has been tested in a single wheel test bed with a wheel driving up to speeds of 0.07 m/s in a wide range of wheel slippage resulting in an average drift of 5 mm/s. A qualitative evaluation of the proposed system against other sources of odometry shows that acoustic and visual methods vulnerabilities do not overlap, which indicates that a system based on acoustic sensors can be a feasible auxiliary source of odometry. The system is able to make a prediction from a single audio frame with a duration of 15 ms in only 2 ms on a user-level commercially available CPU.

As robotic navigation techniques in perception and planning advance, mobile robots increasingly venture into off-road environments involving complex traversability. However, selecting suitable planning methods remains a challenge due to their algorithmic diversity, as each offers unique benefits. To aid in algorithm design, we introduce BenchNav, an open-source PyTorch-based simulation platform for benchmarking off-road navigation with uncertain traversability. Built upon Gymnasium, BenchNav provides three key features: 1) a data generation pipeline for preparing synthetic natural environments, 2) built-in machine learning models for traversability prediction, and 3) consistent execution of path and motion planning across different algorithms. We show BenchNav's versatility through simulation examples in off-road environments, employing three representative planning algorithms from different domains. https://github.com/masafumiendo/benchnav

Cooperative exploration by unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) is efficient for long-duration surveys and wider exploration areas. We propose a rendezvous point determination and path planning method that reduces the travel distance between survey points in a UAV-UGV cooperative system, in which the UAV can recharge itself by landing on the UGV. In this study, Bidirectional Informed Rapidly-exploring Random Tree* that selectively extends paths for UAV and UGV in a specific direction is developed. Simulation study in three different terrains confirms that this method can reduce the travel distance by 5–7% compared to existing methods.

The prototypical diluted magnetic semiconductor Cd1-xMnxTe is a spin glass (x<0.6) or an antiferromagnet (x>0.6), but becomes ferromagnetic upon doping with a small amount of Cr atoms substituting for Mn. In order to investigate the origin of the ferromagnetism in Cd1-x-yMnxCryTe, we have studied its element specific magnetic properties by x-ray absorption spectroscopy (XAS) and x-ray magnetic circular dichroism (XMCD) at the Cr and Mn L2,3 edges. Thin films were grown by molecular beam epitaxy with a fixed Mn content of x = 0.2 and varying Cr content in the range of y = 0 - 0.04. Measured XAS and XMCD spectra indicate that both Cr and Mn atoms are divalent and that the ferromagnetic or superparamagnetic components of Cr and Mn are aligned in the same directions. The magnetization of Mn increases with increasing Cr content. These results can be explained if ferromagnetic interaction exists between neighboring Mn and Cr ions although interaction between Mn atoms is largely antiferromagnetic. We conclude that each ferromagnetic or superparamagnetic cluster consists of ferromagnetically coupled several Cr and a much larger number of Mn ions.

Traversability assessment of deformable terrain is vital for safe rover navigation on planetary surfaces. Machine learning (ML) is a powerful tool for traversability prediction but faces predictive uncertainty. This uncertainty leads to prediction errors, increasing the risk of wheel slips and immobilization for planetary rovers. To address this issue, we integrate principal approaches to uncertainty handling -- quantification, exploitation, and adaptation -- into a single learning and planning framework for rover navigation. The key concept is \emph{deep probabilistic traversability}, forming the basis of an end-to-end probabilistic ML model that predicts slip distributions directly from rover traverse observations. This probabilistic model quantifies uncertainties in slip prediction and exploits them as traversability costs in path planning. Its end-to-end nature also allows adaptation of pre-trained models with in-situ traverse experience to reduce uncertainties. We perform extensive simulations in synthetic environments that pose representative uncertainties in planetary analog terrains. Experimental results show that our method achieves more robust path planning under novel environmental conditions than existing approaches.

Analyzing the mobility of wheeled rovers on loose sand in low-gravity environments remains a significant challenge. Among several experimental techniques, such as parabolic flight and reduced-weight tests, granular scaling laws (GSL) have recently been proposed to predict wheel mobility under low-gravity conditions via earth-gravity tests. Although the GSL accurately predicts wheel mobility on flat terrain in low-gravity environments, its capability to predict wheel mobility on slopes in such environments still needs to be verified. This study developed a GSL and investigated its accuracy in predicting wheel mobility on slopes in low-gravity environments. The discrete element method (DEM) was utilized to test wheel mobility at various slope angles under Earth's gravity. Subsequently, by applying a multiple scaling function, the GSL converted the results from the Earth-gravity tests to predict wheel mobility under lunar gravity. The GSL-based predictions were compared with DEM simulations conducted under lunar gravity conditions. The results indicated that the wheel mobility under lunar gravity predicted by the GSL closely corresponded to that calculated via the DEM. These findings indicate that the GSL can accurately predict wheeled-rover mobility on slopes in low-gravity environments.