Legged robots that can operate autonomously in remote and hazardous environments will greatly increase opportunities for exploration into underexplored areas. Exteroceptive perception is crucial for fast and energy-efficient locomotion: Perceiving the terrain before making contact with it enables planning and adaptation of the gait ahead of time to maintain speed and stability. However, using exteroceptive perception robustly for locomotion has remained a grand challenge in robotics. Snow, vegetation, and water visually appear as obstacles on which the robot cannot step or are missing altogether due to high reflectance. In addition, depth perception can degrade due to difficult lighting, dust, fog, reflective or transparent surfaces, sensor occlusion, and more. For this reason, the most robust and general solutions to legged locomotion to date rely solely on proprioception. This severely limits locomotion speed because the robot has to physically feel out the terrain before adapting its gait accordingly. Here, we present a robust and general solution to integrating exteroceptive and proprioceptive perception for legged locomotion. We leverage an attention-based recurrent encoder that integrates proprioceptive and exteroceptive input. The encoder is trained end to end and learns to seamlessly combine the different perception modalities without resorting to heuristics. The result is a legged locomotion controller with high robustness and speed. The controller was tested in a variety of challenging natural and urban environments over multiple seasons and completed an hour-long hike in the Alps in the time recommended for human hikers.

Robots working in natural, urban, and industrial settings need to be able to navigate challenging environments. In this paper, we present a motion planner for the perceptive rough-terrain locomotion with quadrupedal robots. The planner finds safe footholds along with collision-free swing-leg motions by leveraging an acquired terrain map. To this end, we present a novel pose optimization approach that enables the robot to climb over significant obstacles. We experimentally validate our approach with the quadrupedal robot ANYmal by autonomously traversing obstacles such steps, inclines, and stairs. The locomotion planner re-plans the motion at every step to cope with disturbances and dynamic environments. The robot has no prior knowledge of the scene, and all mapping, state estimation, control, and planning is performed in real-time onboard the robot.

BACKGROUND AND PURPOSEHydrogen sulfide (H2S), generated by enzymes such as cystathionine-g-lyase (CSE) from L-cysteine, facilitates pain signals by activating the Cav3.2 T-type Ca 2+ channels. Here, we assessed the involvement of the CSE/H2S/Cav3.2 pathway in cystitis-related bladder pain. EXPERIMENTAL APPROACHCystitis was induced by i.p. administration of cyclophosphamide in mice. Bladder pain-like nociceptive behaviour was observed and referred hyperalgesia was evaluated using von Frey filaments. Phosphorylation of ERK in the spinal dorsal horn was determined immunohistochemically following intravesical administration of NaHS, an H2S donor. KEY RESULTSCyclophosphamide caused cystitis-related symptoms including increased bladder weight, accompanied by nociceptive changes (bladder pain-like nociceptive behaviour and referred hyperalgesia). Pretreatment with DL-propargylglycine, an inhibitor of CSE, abolished the nociceptive changes and partly prevented the increased bladder weight. CSE protein in the bladder was markedly up-regulated during development of cystitis. Mibefradil or NNC 55-0396, blockers of T-type Ca 2+ channels, administered after the symptoms of cystitis appeared, reversed the nociceptive changes. Further, silencing of Cav3.2 protein by repeated intrathecal administration of mouse Cav3.2-targeting antisense oligodeoxynucleotides also significantly attenuated the nociceptive changes, but not the increased bladder weight. Finally, the number of cells staining positive for phospho-ERK was increased in the superficial layer of the L6 spinal cord after intravesical administration of NaHS, an effect inhibited by NNC 55-0396. CONCLUSION AND IMPLICATIONSEndogenous H2S, generated by up-regulated CSE, caused bladder pain and referred hyperalgesia through the activation of Cav3.2 channels, one of the T-type Ca 2+ channels, in mice with cyclophosphamide-induced cystitis. Abbreviations

Given the previous evidence for involvement of prostanoid EP1 receptors in facilitation of the bladder afferent nerve activity and micturition reflex, the present study investigated the effect of ONO-8130, a selective EP1 receptor antagonist, on cystitis-related bladder pain in mice. Cystitis in mice was produced by intraperitoneal administration of cyclophosphamide at 300mg/kg. Bladder pain-like nociceptive behavior and referred hyperalgesia were assessed in conscious mice. Phosphorylation of extracellular signal-regulated kinase (ERK) in the L6 spinal cord was determined by immunohistochemistry in anesthetized mice. Cyclophosphamide treatment caused bladder pain-like nociceptive behavior and referred hyperalgesia accompanying cystitis symptoms, including increased bladder weight and vascular permeability and upregulation of cyclooxygenase-2 in the bladder tissue. Oral preadministration of ONO-8130 at 0.3-30 mg/kg strongly prevented both the bladder pain-like behavior and referred hyperalgesia in a dose-dependent manner, but had slight effect on the increased bladder weight and vascular permeability. Oral ONO-8130 at 30 mg/kg also reversed the established cystitis-related bladder pain. Intravesical administration of prostaglandin E2 caused prompt phosphorylation of ERK in the L6 spinal cord, an effect blocked by ONO-8130. Our findings strongly suggest that the prostaglandin E2/EP1 system participates in processing of cystitis-related bladder pain, and that EP1 antagonists including ONO-8130 are useful for treatment of bladder pain, particularly in interstitial cystitis. Prostaglandin E2 contributes to cystitis-related bladder pain via EP1 receptors in mice, indicating possible therapeutic usefulness of selective EP1 antagonists.

This article presents the core technologies and deployment strategies of Team CERBERUS that enabled our winning run in the DARPA Subterranean Challenge finals. CERBERUS is a robotic system-of-systems involving walking and flying robots presenting resilient autonomy, as well as mapping and navigation capabilities to explore complex underground environments.