Wheeled-legged robots are an attractive solution for versatile locomotion in challenging terrain. They combine the speed and efficiency of wheels with the ability of legs to traverse challenging terrain. In this paper, we present a trajectory optimization formulation for wheeled-legged robots that optimizes over the base and wheels' positions and forces and takes into account the terrain information while computing the plans. This enables us to find optimal driving motions over challenging terrain. The robot is modeled as a single rigid-body, which allows us to plan complex motions and still keep a low computational complexity to solve the optimization quickly. The terrain map, together with the use of a stability constraint, allows the optimizer to generate feasible motions that cannot be discovered without taking the terrain information into account. The optimization is formulated as a Nonlinear Programming (NLP) problem and the reference motions are tracked by a hierarchical whole-body controller that computes the torque actuation commands for the robot. The trajectories have been experimentally verified on quadrupedal robot ANYmal equipped with non-steerable torque-controlled wheels. Our trajectory optimization framework enables wheeled quadrupedal robots to drive over challenging terrain, e.g., steps, slopes, stairs, while negotiating these obstacles with dynamic motions.

We describe an optimization-based framework to perform complex locomotion skills for robots with legs and wheels. The generation of complex motions over a long-time horizon often requires offline computation due to current computing constraints and is mostly accomplished through trajectory optimization (TO). In contrast, model predictive control (MPC) focuses on the online computation of trajectories, robust even in the presence of uncertainty, albeit mostly over shorter time horizons and is prone to generating nonoptimal solutions over the horizon of the task’s goals. Our article’s contributions overcome this trade-off by combining offline motion libraries and online MPC, uniting a complex, long-time horizon plan with reactive, short-time horizon solutions. We start from offline trajectories that can be, for example, generated by TO or sampling-based methods. Also, multiple offline trajectories can be composed out of a motion library into a single maneuver. We then use these offline trajectories as the cost for the online MPC, allowing us to smoothly blend between multiple composed motions even in the presence of discontinuous transitions. The MPC optimizes from the measured state, resulting in feedback control, which robustifies the task’s execution by reacting to disturbances and looking ahead at the offline trajectory. With our contribution, motion designers can choose their favorite method to iterate over behavior designs offline without tuning robot experiments, enabling them to author new behaviors rapidly. Our experiments demonstrate complex and dynamic motions on our traditional quadrupedal robot ANYmal and its roller-walking version. Moreover, the article’s findings contribute to evaluating five planning algorithms.

In digital farming, the use of technology to increase agricultural production through automated tasks has recently integrated the development of AgBots for more reliable data collection using autonomous navigation. These AgBots are equipped with various sensors such as GNSS, cameras, and LiDAR, but these sensors can be prone to limitations such as low accuracy for under-canopy navigation with GNSS, sensitivity to outdoor lighting and platform vibration with cameras, and LiDAR occlusion issues. In order to address these limitations and ensure robust autonomous navigation, this paper presents a sensor selection methodology based on the identification of environmental conditions using sensor data. Through the extraction of features from GNSS, images, and point clouds, we are able to determine the feasibility of using each sensor and create a selection vector indicating its viability. Our results demonstrate that the proposed methodology effectively selects between the use of cameras or LiDAR within crops and GNSS outside of crops, at least 87% of the time. The main problem found is that, in the transition from inside to outside and from outside to inside the crop, GNSS features take 20 s to adapt. We compare a variety of classification algorithms in terms of performance and computational cost and the results show that our method has higher performance and lower computational cost. Overall, this methodology allows for the low-cost selection of the most suitable sensor for a given agricultural environment.

No abstract

No abstract