racket op• 数据种类:table × table × table → table
• 调用: import fullseye as fs; fs.ledger.rally_simulate(bp: 'dict', rp: 'dict', tp: 'dict', *, perceive=None, strategies=None, retries: 'int' = 6, x_racket: 'float' = 1.55, hit_height=(0.05, 0.45), T_flight: 'float' = 0.42, target_x: 'float' = 0.75, max_hits: 'int' = 60, dt: 'float' = 0.0002, fps: 'float' = 100.0, seed: 'int' = 0, y_jitter: 'float' = 0.25, plan_spin: 'bool' = True, table_ip=None, replan_every: 'int' = 5) -> 'dict'(要直接调用实现,import racket; racket.rally_simulate(bp: 'dict', rp: 'dict', tp: 'dict', *, perceive=None, strategies=None, retries: 'int' = 6, x_racket: 'float' = 1.55, hit_height=(0.05, 0.45), T_flight: 'float' = 0.42, target_x: 'float' = 0.75, max_hits: 'int' = 60, dt: 'float' = 0.0002, fps: 'float' = 100.0, seed: 'int' = 0, y_jitter: 'float' = 0.25, plan_spin: 'bool' = True, table_ip=None, replan_every: 'int' = 5) -> 'dict';从台账取用则 opsdrive.get("rally_simulate"))
2 支球拍(位于 x = −x_r 与 +x_r，拍面朝向对方)相互对打。
每帧(1/fps)接球方的感知 perceive(t, p, v, ω) 返回球的状态,据此用运动方程 + 球台弹跳预测到达自己平面 x = ±x_r 的到达点和时刻,并移动球拍(有上限)。球到达平面时,瞄准在 T_flight 内落在对方球台 (∓target_x, y*) 的速度击球(y* 为随机数;若 plan_spin 则在瞄准时考虑击球后的旋转)。strategies = (side 0 的策略, side 1 的策略)(None 则双方均为 strategy_feeder)。若 retries > 0,则击球前用 shot_is_legal 预判,会失误的瞄准被回滚,改用别的瞄准(重新抽取策略)重来。返回值 {"hits", "end_reason", "shots": [...], "t", "p" (N,3), "racket_pos" (N,2,3), "hit_times", "rewinds"}。
空输入(bp 为空的 dict)会引发 ValueError。
• 示例数据目录(下载 URL / 许可证) —— 2-D 用 skimage.data(BSD/公有领域)加合成图,3-D 给出真实数据源(Stanford/PDS 等)的下载 URL。
• 算子来历与参考文献 —— 该算子族所依据的研究/方法出处。
• 算法的正典(作者、年份)与用途见上面的族使用指南。
• poc_ball_bounce — py -3.11 examples/poc_ball_bounce.py
• poc_table_tennis_rally_loop — py -3.11 examples/poc_table_tennis_rally_loop.py
table 作为输入)course_layout · course_occupancy · course_contains · world_build · world_camera · world_move · lidar_scan · rss_longitudinal_same
racket)racket_params · racket_impact · racket_hit_check · aim_velocity · racket_plan · racket_move · strategy_attacker · strategy_feeder
*Provenance: racket.py — DRIVE 算子登记表。本条目由 tools/opdocs.py md 自动生成(请勿手工编辑)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.