In this paper we propose a new method of encoding social norms and rules into sampling-based motion planners. Inspired from the social force model (SFM), we modify and use it as a social intention model (SIM) to reshape the motion primitives (MP) of the rapidly-exploring random tree (RRT) motion planner for the socially aware robot navigation. We also introduce a new benchmark for evaluating social planning performance, so called as the social effort index (SEI). The experimental results show that the socially-guided motion primitives-based RRT increases safe and social interactions between the robot and human agents about 50% compared to the typical RRT-embedded MP (RRT-MP) in human populated environments.

Engineering students in a project-based curriculum are expected to learn and apply the engineering design process to their course embedded projects. Practice of embedded skills typically occurs through embarking on a new project context provided by an instructor. It is a rare occurrence for students to participate in experiences that break-up the process into smaller chunks providing low-pressure instances to practice. The following work-in-progress describes a practice-then-apply scaffolding approach to teaching engineering design. The engineering design process was broken down into three phases: (1) discovery and ideation, (2) concept development and selection, and (3) realization and experimentation. Each phase was presented to students via a mini-project that inserted students into the design process at various stages of the design. The mini-projects afforded students with opportunities to practice and expand their understanding without always having to start at the beginning. A final project embedded throughout the course provided students with the opportunity to apply what they had learned in the mini-projects. This scaffolded approach methodically slowed the process providing a unique design learning experience with explicit design activities.

Optimal heuristic searches such as A* search are commonly used for low-dimensional planning such as 2D path finding. These algorithms however, typically do not scale well to high-dimensional planning problems such as motion planning for robotic arms, computing motion trajectories for non-holonomic robotic vehicles and motion synthesis for humanoid characters. A recently developed randomized version of A* search, called R* search, scales to higher-dimensional planning problems by trading off deterministic optimality guarantees of A* for probabilistic sub-optimality guarantees. In this paper, we show that in addition to its scalability, R* lends itself well to a parallel implementation. In particular, we demonstrate how R* can be implemented on the GPU. On the theoretical side, the GPU version of R*, called R*GPU, preserves all the theoretical properties of R* including its probabilistic bounds on sub-optimality. On the experimental side, we show that R*GPU consistently produces lower cost solutions, scales better in terms of memory, and runs faster than R*. These results hold for both motion planning for a 6DOF robot arm planar as well as 2D path finding.