Motion Planning for Actively Reconﬁgurable Mobile Robots in Search and Rescue Scenarios Michael Brunner, Bernd Br¨ uggemann and Dirk Schulz Fraunhofer Institute for Communication, Information Processing and Ergonomics FKIE, Wachtberg, Germany Email: {michael.brunner, bernd.brueggemann, dirk.schulz}@fkie.fraunhofer.de Abstract — In disaster scenarios, mobile robots can be employed in hazardous environments where it is too dangerous for human rescuers. Robotic systems can assist rescue personnel as they can be used to explore those inaccessible areas and to assess the situation. Tracked platforms with actuators have been proven to be well suited for such deployments because they are agile enough to overcome quite challenging terrain. A very demanding task for operators is the navigation of the robotic system in complex disaster environments. Hence, an important capability of future systems for search and rescue missions is autonomous navigation in disaster scenarios. In this paper we introduce a two-phase motion planning algorithm for tracked robots with actively controlled actuators to ﬁnd a fast and stable path to a user speciﬁed goal. In the ﬁrst phase, we generate an initial path considering the platform’s operating limits and the terrain roughness. In the second phase, we limit the search space to the area around the initial path and reﬁne the preliminary solution accounting for the complete robot state including actuators and the robot’s stability and traction. A main distinction of our method is that it does not rely on a previous classiﬁcation of the terrain, thus, can be applied to a variety of environments. We present experiments evaluating our algorithm in simulation and in two real-world scenarios to demonstrate the validity and feasibility of our approach. Keywords: autonomy, motion planning, rough terrain, re- conﬁgurable chassis, mobile robot, actuators I. I NTRODUCTION The goal in search and rescue robotics is two-fold; to extend the capabilities of human rescue personnel and to improve their safety. Usually, a teleoperated system is deployed, con- trolled with or without line-of-sight. The mobile system is used to initially explore the environment and assess the situation. Controlling a robotic system in complex environments like disaster scenarios is a challenging task for operators, especially because they are unable to use their natural human senses and are constrained to the limited and unusual views provided by the robot’s sensors. In addition to judging the situation and searching for victims, the operator must ﬁnd valid paths for the robot. As achieving full autonomy of search and rescue robots is very hard due to the complexity of those mission, many assistance functions for teleoperated systems [1] or for semi- autonomous operation [2], [3] have been proposed in order to reduce the cognitive burden on the operator. These methods in- clude improved graphical user interfaces, processing of sensor data, information extraction algorithms and victim detection strategies. Fig. 1: Method overview: First, a preprocessing step which analyses the environment, followed by a preliminary planning step to ﬁnd an initial path and a detailed motion planning step which reﬁnes the initial solution in rough areas using the actuators. In ﬂat areas, a default conﬁguration is applied. Navigating a mobile robot through complex environments under the previously described conditions and searching for a path is very demanding for an operator. Especially because the path should not damage the robot, should avoid deadlocks and yet should still serve the mission. The operator must consider many different aspects when driving through rough terrain and over obstacles. The stability of the robot is critical as the robot may fall over. Inertia and momentum become increasingly important if a robot is operated close to its limits. Additionally, varying contact points of the mobile system change its behavior to actuator and driving commands. In this paper, we present a two-phase motion planning algorithm for tracked actively reconﬁgurable robots to ﬁnd a fast and stable path to a user speciﬁed goal in rough terrain (Fig. 1). We ﬁrst ﬁnd an initial path quickly using the robot’s operating limits while neglecting its actuators. We then restrict the high dimensional search space, which includes the actuators, to the areas around those initial path segments which lead through rough regions of the environment. Thus, we speed up planning while considering the actuators, the stability of the system and its traction. Our method does not rely on terrain classiﬁcations or ﬁxed motion sequences. Hence, it can be applied to rough, unstructured outdoor environments as well as obstacles in urban surroundings (e.g. stairs). We consider our method to be the global planning compo- nent of a robotic system which provides a plan. Followed by a feedback controller which takes care of the plan execution