Search for: All records

Recent advances in reinforcement learning (RL) have demonstrated impressive capabilities in complex decision-making tasks. This progress raises a natural question: how do these artificial systems compare to biological agents, which have been shaped by millions of years of evolution? To help answer this question, we undertake a comparative study of biological mice and RL agents in a predator-avoidance maze environment. Through this analysis, we identify a striking disparity: RL agents consistently demonstrate a lack of self-preservation instinct, readily risking ``death'' for marginal efficiency gains. These risk-taking strategies are in contrast to biological agents, which exhibit sophisticated risk-assessment and avoidance behaviors. Towards bridging this gap between the biological and artificial, we propose two novel mechanisms that encourage more naturalistic risk-avoidance behaviors in RL agents. Our approach leads to the emergence of naturalistic behaviors, including strategic environment assessment, cautious path planning, and predator avoidance patterns that closely mirror those observed in biological systems. 

We present planning and control techniques for non-periodic locomotion tasks specified by temporal logic in rough cluttered terrains. Our planning approach is based on a discrete set of motion primitives for the center of mass (CoM) of a general bipedal robot model. A deterministic shortest path problem is solved over the Bu ̈chi automaton of the temporal logic task specification, composed with the graph of CoM keyframe states generated by the motion primitives. A low-level controller based on quadratic programming is proposed to track the resulting CoM and foot trajectories. We demonstrate dynamically stable, non-periodic locomotion of a kneed compass gait bipedal robot satisfying complex task specifications.