feat(training): release V0.9.1 避障训练与基础策略迁移

2026-09-08 10:50:13 +08:00
parent fa5485049a
commit 438e56bcc8
113 changed files with 15027 additions and 539 deletions
@@ -0,0 +1,209 @@
+import { expect, test } from '@playwright/test';
+import { readFileSync, writeFileSync } from 'node:fs';
+import { resolve } from 'node:path';
+
+// Opt-in real source-derived actor, not a constant fixture. Run against the Vite dev server.
+const policyPath = process.env.GO2_PRETRAINED_NAV_POLICY;
+test('真实warm-start策略连续行走超过20秒并换目标，不重载策略或自动启用', async ({
+  page,
+}, testInfo) => {
+  test.skip(
+    !policyPath,
+    '设置GO2_PRETRAINED_NAV_POLICY为真实47→81/97 warm-start导出，并运行Vite dev :4173',
+  );
+  const xml = readFileSync(
+    resolve('training_server/rl/src/assets/robots/unitree_go2/xmls/go2.xml'),
+    'utf8',
+  )
+    .replace(/<mesh\b[^>]*\/>/g, '')
+    .replace(/<geom\b[^>]*\bmesh="[^"]*"[^>]*\/>/g, '');
+  const policy = Array.from(readFileSync(policyPath!));
+  await page.goto(process.env.GO2_PRETRAINED_DEV_URL ?? 'http://127.0.0.1:4174');
+  const result = await page.evaluate(
+    async ({ xml, policy }) => {
+      const adapterPath = '/src/simulation/PhysicsAdapter.ts';
+      const deploymentPath = '/src/rl/deployment.ts';
+      const navigationPath = '/src/rl/tasks/go2ObstacleAvoidance.ts';
+      const { MainThreadPhysicsAdapter } = (await import(
+        adapterPath
+      )) as typeof import('../src/simulation/PhysicsAdapter');
+      const { readPolicyDeployment } = (await import(
+        deploymentPath
+      )) as typeof import('../src/rl/deployment');
+      const { obstacleNavigation } = (await import(
+        navigationPath
+      )) as typeof import('../src/rl/tasks/go2ObstacleAvoidance');
+      const bytes = new Uint8Array(policy),
+        data = new TextEncoder().encode(xml);
+      const deployment = readPolicyDeployment(bytes)!;
+      const adapter = new MainThreadPhysicsAdapter();
+      try {
+        await adapter.load(
+          {
+            id: 'real-pretrained',
+            name: 'go2',
+            files: [
+              { path: 'go2.xml', data, size: data.length, source: 'file', mimeType: 'text/xml' },
+            ],
+            entries: [{ path: 'go2.xml', format: 'mjcf', label: 'Go2' }],
+            maps: [],
+            totalBytes: data.length,
+          },
+          'go2.xml',
+          {
+            trainingDeployment: deployment,
+            trainingPolicy: { data: bytes, path: 'user-warmstart.onnx' },
+          },
+        );
+        const session = adapter.session!;
+        const originalPolicy = (session as unknown as { rlPolicy: object }).rlPolicy;
+        const initial = adapter.snapshot()!;
+        const samples: {
+          time: number;
+          x: number;
+          y: number;
+          z: number;
+          enabled?: boolean;
+          error?: string;
+        }[] = [];
+        const run = async (seconds: number) => {
+          const target = Number(session.data.time) + seconds;
+          while (Number(session.data.time) < target) {
+            for (let i = 0; i < 10; i++) adapter.singleStep();
+            await new Promise((resolve) => setTimeout(resolve, 0));
+            if (!adapter.snapshot()?.rlPolicy?.enabled) break;
+          }
+          const s = adapter.snapshot()!;
+          samples.push({
+            time: s.time,
+            x: s.qpos[0],
+            y: s.qpos[1],
+            z: s.qpos[2],
+            enabled: s.rlPolicy?.enabled,
+            error: s.rlPolicy?.error,
+          });
+          return s;
+        };
+        const before = await run(21);
+        const navBefore = obstacleNavigation(
+          before.qpos.slice(0, 3),
+          before.qpos.slice(3, 7),
+          before.rlPolicy!.navigation!.target,
+          deployment.terrain!.size,
+          deployment.navigation?.speed,
+        );
+        // Nearby diagonal target exercises both heading and velocity command; no pose teleport.
+        adapter.setNavigationTarget([before.qpos[0] - 2, before.qpos[1] + 2]);
+        const changed = adapter.snapshot()!;
+        const navChanged = obstacleNavigation(
+          changed.qpos.slice(0, 3),
+          changed.qpos.slice(3, 7),
+          changed.rlPolicy!.navigation!.target,
+          deployment.terrain!.size,
+          deployment.navigation?.speed,
+        );
+        const after = await run(4);
+        adapter.setRLPolicyEnabled(false);
+        adapter.setNavigationTarget([0, 0]);
+        const disabled = adapter.snapshot()!;
+        return {
+          initial: { time: initial.time, qpos: initial.qpos },
+          before: {
+            time: before.time,
+            qpos: before.qpos,
+            ctrl: before.ctrl,
+            status: before.rlPolicy,
+          },
+          after: { time: after.time, qpos: after.qpos, ctrl: after.ctrl, status: after.rlPolicy },
+          samples,
+          navBefore,
+          navChanged,
+          sameSession: adapter.session === session,
+          samePolicy: (session as unknown as { rlPolicy: object }).rlPolicy === originalPolicy,
+          heldControlUnchangedOnTarget: changed.ctrl.every((v, i) => v === before.ctrl[i]),
+          countBeforeTarget: before.rlPolicy!.inferenceCount,
+          countAfterTarget: changed.rlPolicy!.inferenceCount,
+          disabledAfterNewTarget: !disabled.rlPolicy!.enabled,
+        };
+      } finally {
+        adapter.dispose();
+      }
+    },
+    { xml, policy },
+  );
+  const evidence = testInfo.outputPath('real-navigation-measurements.json');
+  writeFileSync(evidence, JSON.stringify(result, null, 2));
+  await testInfo.attach('real-navigation-measurements.json', {
+    path: evidence,
+    contentType: 'application/json',
+  });
+  expect(result.before.time).toBeGreaterThan(20);
+  expect(result.before.status?.enabled).toBe(true);
+  expect(result.after.status?.enabled).toBe(true);
+  expect(
+    Math.hypot(
+      result.before.qpos[0] - result.initial.qpos[0],
+      result.before.qpos[1] - result.initial.qpos[1],
+    ),
+  ).toBeGreaterThan(0.5);
+  expect(result.navChanged.command).not.toEqual(result.navBefore.command);
+  expect(result.navChanged.targetError).not.toEqual(result.navBefore.targetError);
+  expect(result.after.ctrl).not.toEqual(result.before.ctrl);
+  expect(result.after.status!.navigation!.distance).toBeLessThan(2 * Math.SQRT2);
+  expect(result.countAfterTarget).toBe(result.countBeforeTarget);
+  expect(result.after.status!.inferenceCount).toBeGreaterThan(result.countBeforeTarget);
+  expect(result.sameSession).toBe(true);
+  expect(result.samePolicy).toBe(true);
+  expect(result.heldControlUnchangedOnTarget).toBe(true);
+  expect(result.disabledAfterNewTarget).toBe(true);
+});
+
+test('用户原始47维ONNX通过普通Flat面板加载，不新增点击导航模式', async ({ page }) => {
+  const source = process.env.GO2_PRETRAINED_FLAT_POLICY;
+  const robotXml = process.env.GO2_PRETRAINED_FLAT_XML;
+  test.skip(!source || !robotXml, '设置原始47维ONNX与同源带actuator的机器人XML路径');
+  const xml = readFileSync(robotXml!, 'utf8')
+    .replace(/<mesh\b[^>]*\/>/g, '')
+    .replace(/<geom\b[^>]*\bmesh="[^"]*"[^>]*\/>/g, '');
+  await page.goto(process.env.GO2_PRETRAINED_DEV_URL ?? 'http://127.0.0.1:4174');
+  await page
+    .locator('input[type="file"]')
+    .first()
+    .setInputFiles({ name: 'go2.xml', mimeType: 'text/xml', buffer: Buffer.from(xml) });
+  await expect(page.getByText('WASM 已加载')).toBeVisible({ timeout: 30_000 });
+  await page.getByRole('tab', { name: '控制台' }).click();
+  const tools = page.getByRole('tabpanel', { name: '控制台' });
+  const section = tools.getByRole('button', { name: /ONNX 策略运行/ });
+  if ((await section.getAttribute('aria-expanded')) === 'false') await section.click();
+  const chooser = page.waitForEvent('filechooser');
+  await tools.getByRole('button', { name: '导入 ONNX' }).click();
+  await (await chooser).setFiles(source!);
+  await expect
+    .poll(
+      async () =>
+        (await tools.getByText('47 / 12').count()) +
+        (await page.getByRole('button', { name: '技术详情', exact: true }).count()),
+    )
+    .toBeGreaterThan(0);
+  const details = page.getByRole('button', { name: '技术详情', exact: true });
+  if (await details.isVisible()) {
+    await details.click();
+    throw new Error((await page.getByRole('alert').allTextContents()).join('\n'));
+  }
+  await expect(tools.getByText('47 / 12')).toBeVisible({ timeout: 30_000 });
+  const enable = tools.getByRole('button', { name: '启用', exact: true });
+  if (await enable.isVisible()) await enable.click();
+  const play = page.getByRole('button', { name: '▶ 播放', exact: true });
+  if (await play.isVisible()) await play.click();
+  await expect
+    .poll(async () =>
+      Number(
+        (await tools.getByText('推理次数', { exact: true }).locator('..').textContent())?.replace(
+          /\D/g,
+          '',
+        ),
+      ),
+    )
+    .toBeGreaterThan(2);
+  await expect(tools.getByRole('button', { name: '设定目标', exact: true })).toHaveCount(0);
+});